{-# LANGUAGE CPP #-} {-# OPTIONS_GHC -Wno-incomplete-uni-patterns #-} {- (c) The GRASP/AQUA Project, Glasgow University, 1993-1998 \section[Specialise]{Stamping out overloading, and (optionally) polymorphism} -} module GHC.Core.Opt.Specialise ( specProgram, specUnfolding ) where #include "HsVersions.h" import GHC.Prelude import GHC.Driver.Session import GHC.Driver.Ppr import GHC.Driver.Config import GHC.Driver.Env import GHC.Tc.Utils.TcType hiding( substTy ) import GHC.Core.Type hiding( substTy, extendTvSubstList ) import GHC.Core.Multiplicity import GHC.Core.Predicate import GHC.Core.Coercion( Coercion ) import GHC.Core.Opt.Monad import qualified GHC.Core.Subst as Core import GHC.Core.Unfold.Make import GHC.Core import GHC.Core.Rules import GHC.Core.Utils ( exprIsTrivial, getIdFromTrivialExpr_maybe , mkCast, exprType ) import GHC.Core.FVs import GHC.Core.TyCo.Rep (TyCoBinder (..)) import GHC.Core.Opt.Arity ( collectBindersPushingCo , etaExpandToJoinPointRule ) import GHC.Builtin.Types ( unboxedUnitTy ) import GHC.Data.Maybe ( mapMaybe, maybeToList, isJust ) import GHC.Data.Bag import GHC.Data.FastString import GHC.Types.Basic import GHC.Types.Unique.Supply import GHC.Types.Unique.DFM import GHC.Types.Name import GHC.Types.Tickish import GHC.Types.Id.Make ( voidArgId, voidPrimId ) import GHC.Types.Var ( isLocalVar ) import GHC.Types.Var.Set import GHC.Types.Var.Env import GHC.Types.Id import GHC.Utils.Monad ( foldlM ) import GHC.Utils.Misc import GHC.Utils.Outputable import GHC.Utils.Panic import GHC.Unit.Module( Module ) import GHC.Unit.Module.ModGuts import GHC.Unit.External {- ************************************************************************ * * \subsection[notes-Specialise]{Implementation notes [SLPJ, Aug 18 1993]} * * ************************************************************************ These notes describe how we implement specialisation to eliminate overloading. The specialisation pass works on Core syntax, complete with all the explicit dictionary application, abstraction and construction as added by the type checker. The existing type checker remains largely as it is. One important thought: the {\em types} passed to an overloaded function, and the {\em dictionaries} passed are mutually redundant. If the same function is applied to the same type(s) then it is sure to be applied to the same dictionary(s)---or rather to the same {\em values}. (The arguments might look different but they will evaluate to the same value.) Second important thought: we know that we can make progress by treating dictionary arguments as static and worth specialising on. So we can do without binding-time analysis, and instead specialise on dictionary arguments and no others. The basic idea ~~~~~~~~~~~~~~ Suppose we have let f = <f_rhs> in <body> and suppose f is overloaded. STEP 1: CALL-INSTANCE COLLECTION We traverse <body>, accumulating all applications of f to types and dictionaries. (Might there be partial applications, to just some of its types and dictionaries? In principle yes, but in practice the type checker only builds applications of f to all its types and dictionaries, so partial applications could only arise as a result of transformation, and even then I think it's unlikely. In any case, we simply don't accumulate such partial applications.) STEP 2: EQUIVALENCES So now we have a collection of calls to f: f t1 t2 d1 d2 f t3 t4 d3 d4 ... Notice that f may take several type arguments. To avoid ambiguity, we say that f is called at type t1/t2 and t3/t4. We take equivalence classes using equality of the *types* (ignoring the dictionary args, which as mentioned previously are redundant). STEP 3: SPECIALISATION For each equivalence class, choose a representative (f t1 t2 d1 d2), and create a local instance of f, defined thus: f@t1/t2 = <f_rhs> t1 t2 d1 d2 f_rhs presumably has some big lambdas and dictionary lambdas, so lots of simplification will now result. However we don't actually *do* that simplification. Rather, we leave it for the simplifier to do. If we *did* do it, though, we'd get more call instances from the specialised RHS. We can work out what they are by instantiating the call-instance set from f's RHS with the types t1, t2. Add this new id to f's IdInfo, to record that f has a specialised version. Before doing any of this, check that f's IdInfo doesn't already tell us about an existing instance of f at the required type/s. (This might happen if specialisation was applied more than once, or it might arise from user SPECIALIZE pragmas.) Recursion ~~~~~~~~~ Wait a minute! What if f is recursive? Then we can't just plug in its right-hand side, can we? But it's ok. The type checker *always* creates non-recursive definitions for overloaded recursive functions. For example: f x = f (x+x) -- Yes I know its silly becomes f a (d::Num a) = let p = +.sel a d in letrec fl (y::a) = fl (p y y) in fl We still have recursion for non-overloaded functions which we specialise, but the recursive call should get specialised to the same recursive version. Polymorphism 1 ~~~~~~~~~~~~~~ All this is crystal clear when the function is applied to *constant types*; that is, types which have no type variables inside. But what if it is applied to non-constant types? Suppose we find a call of f at type t1/t2. There are two possibilities: (a) The free type variables of t1, t2 are in scope at the definition point of f. In this case there's no problem, we proceed just as before. A common example is as follows. Here's the Haskell: g y = let f x = x+x in f y + f y After typechecking we have g a (d::Num a) (y::a) = let f b (d'::Num b) (x::b) = +.sel b d' x x in +.sel a d (f a d y) (f a d y) Notice that the call to f is at type type "a"; a non-constant type. Both calls to f are at the same type, so we can specialise to give: g a (d::Num a) (y::a) = let f@a (x::a) = +.sel a d x x in +.sel a d (f@a y) (f@a y) (b) The other case is when the type variables in the instance types are *not* in scope at the definition point of f. The example we are working with above is a good case. There are two instances of (+.sel a d), but "a" is not in scope at the definition of +.sel. Can we do anything? Yes, we can "common them up", a sort of limited common sub-expression deal. This would give: g a (d::Num a) (y::a) = let +.sel@a = +.sel a d f@a (x::a) = +.sel@a x x in +.sel@a (f@a y) (f@a y) This can save work, and can't be spotted by the type checker, because the two instances of +.sel weren't originally at the same type. Further notes on (b) * There are quite a few variations here. For example, the defn of +.sel could be floated outside the \y, to attempt to gain laziness. It certainly mustn't be floated outside the \d because the d has to be in scope too. * We don't want to inline f_rhs in this case, because that will duplicate code. Just commoning up the call is the point. * Nothing gets added to +.sel's IdInfo. * Don't bother unless the equivalence class has more than one item! Not clear whether this is all worth it. It is of course OK to simply discard call-instances when passing a big lambda. Polymorphism 2 -- Overloading ~~~~~~~~~~~~~~ Consider a function whose most general type is f :: forall a b. Ord a => [a] -> b -> b There is really no point in making a version of g at Int/Int and another at Int/Bool, because it's only instantiating the type variable "a" which buys us any efficiency. Since g is completely polymorphic in b there ain't much point in making separate versions of g for the different b types. That suggests that we should identify which of g's type variables are constrained (like "a") and which are unconstrained (like "b"). Then when taking equivalence classes in STEP 2, we ignore the type args corresponding to unconstrained type variable. In STEP 3 we make polymorphic versions. Thus: f@t1/ = /\b -> <f_rhs> t1 b d1 d2 We do this. Dictionary floating ~~~~~~~~~~~~~~~~~~~ Consider this f a (d::Num a) = let g = ... in ...(let d1::Ord a = Num.Ord.sel a d in g a d1)... Here, g is only called at one type, but the dictionary isn't in scope at the definition point for g. Usually the type checker would build a definition for d1 which enclosed g, but the transformation system might have moved d1's defn inward. Solution: float dictionary bindings outwards along with call instances. Consider f x = let g p q = p==q h r s = (r+s, g r s) in h x x Before specialisation, leaving out type abstractions we have f df x = let g :: Eq a => a -> a -> Bool g dg p q = == dg p q h :: Num a => a -> a -> (a, Bool) h dh r s = let deq = eqFromNum dh in (+ dh r s, g deq r s) in h df x x After specialising h we get a specialised version of h, like this: h' r s = let deq = eqFromNum df in (+ df r s, g deq r s) But we can't naively make an instance for g from this, because deq is not in scope at the defn of g. Instead, we have to float out the (new) defn of deq to widen its scope. Notice that this floating can't be done in advance -- it only shows up when specialisation is done. User SPECIALIZE pragmas ~~~~~~~~~~~~~~~~~~~~~~~ Specialisation pragmas can be digested by the type checker, and implemented by adding extra definitions along with that of f, in the same way as before f@t1/t2 = <f_rhs> t1 t2 d1 d2 Indeed the pragmas *have* to be dealt with by the type checker, because only it knows how to build the dictionaries d1 and d2! For example g :: Ord a => [a] -> [a] {-# SPECIALIZE f :: [Tree Int] -> [Tree Int] #-} Here, the specialised version of g is an application of g's rhs to the Ord dictionary for (Tree Int), which only the type checker can conjure up. There might not even *be* one, if (Tree Int) is not an instance of Ord! (All the other specialision has suitable dictionaries to hand from actual calls.) Problem. The type checker doesn't have to hand a convenient <f_rhs>, because it is buried in a complex (as-yet-un-desugared) binding group. Maybe we should say f@t1/t2 = f* t1 t2 d1 d2 where f* is the Id f with an IdInfo which says "inline me regardless!". Indeed all the specialisation could be done in this way. That in turn means that the simplifier has to be prepared to inline absolutely any in-scope let-bound thing. Again, the pragma should permit polymorphism in unconstrained variables: h :: Ord a => [a] -> b -> b {-# SPECIALIZE h :: [Int] -> b -> b #-} We *insist* that all overloaded type variables are specialised to ground types, (and hence there can be no context inside a SPECIALIZE pragma). We *permit* unconstrained type variables to be specialised to - a ground type - or left as a polymorphic type variable but nothing in between. So {-# SPECIALIZE h :: [Int] -> [c] -> [c] #-} is *illegal*. (It can be handled, but it adds complication, and gains the programmer nothing.) SPECIALISING INSTANCE DECLARATIONS ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Consider instance Foo a => Foo [a] where ... {-# SPECIALIZE instance Foo [Int] #-} The original instance decl creates a dictionary-function definition: dfun.Foo.List :: forall a. Foo a -> Foo [a] The SPECIALIZE pragma just makes a specialised copy, just as for ordinary function definitions: dfun.Foo.List@Int :: Foo [Int] dfun.Foo.List@Int = dfun.Foo.List Int dFooInt The information about what instance of the dfun exist gets added to the dfun's IdInfo in the same way as a user-defined function too. Automatic instance decl specialisation? ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Can instance decls be specialised automatically? It's tricky. We could collect call-instance information for each dfun, but then when we specialised their bodies we'd get new call-instances for ordinary functions; and when we specialised their bodies, we might get new call-instances of the dfuns, and so on. This all arises because of the unrestricted mutual recursion between instance decls and value decls. Still, there's no actual problem; it just means that we may not do all the specialisation we could theoretically do. Furthermore, instance decls are usually exported and used non-locally, so we'll want to compile enough to get those specialisations done. Lastly, there's no such thing as a local instance decl, so we can survive solely by spitting out *usage* information, and then reading that back in as a pragma when next compiling the file. So for now, we only specialise instance decls in response to pragmas. SPITTING OUT USAGE INFORMATION ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ To spit out usage information we need to traverse the code collecting call-instance information for all imported (non-prelude?) functions and data types. Then we equivalence-class it and spit it out. This is done at the top-level when all the call instances which escape must be for imported functions and data types. *** Not currently done *** Partial specialisation by pragmas ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ What about partial specialisation: k :: (Ord a, Eq b) => [a] -> b -> b -> [a] {-# SPECIALIZE k :: Eq b => [Int] -> b -> b -> [a] #-} or even {-# SPECIALIZE k :: Eq b => [Int] -> [b] -> [b] -> [a] #-} Seems quite reasonable. Similar things could be done with instance decls: instance (Foo a, Foo b) => Foo (a,b) where ... {-# SPECIALIZE instance Foo a => Foo (a,Int) #-} {-# SPECIALIZE instance Foo b => Foo (Int,b) #-} Ho hum. Things are complex enough without this. I pass. Requirements for the simplifier ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ The simplifier has to be able to take advantage of the specialisation. * When the simplifier finds an application of a polymorphic f, it looks in f's IdInfo in case there is a suitable instance to call instead. This converts f t1 t2 d1 d2 ===> f_t1_t2 Note that the dictionaries get eaten up too! * Dictionary selection operations on constant dictionaries must be short-circuited: +.sel Int d ===> +Int The obvious way to do this is in the same way as other specialised calls: +.sel has inside it some IdInfo which tells that if it's applied to the type Int then it should eat a dictionary and transform to +Int. In short, dictionary selectors need IdInfo inside them for constant methods. * Exactly the same applies if a superclass dictionary is being extracted: Eq.sel Int d ===> dEqInt * Something similar applies to dictionary construction too. Suppose dfun.Eq.List is the function taking a dictionary for (Eq a) to one for (Eq [a]). Then we want dfun.Eq.List Int d ===> dEq.List_Int Where does the Eq [Int] dictionary come from? It is built in response to a SPECIALIZE pragma on the Eq [a] instance decl. In short, dfun Ids need IdInfo with a specialisation for each constant instance of their instance declaration. All this uses a single mechanism: the SpecEnv inside an Id What does the specialisation IdInfo look like? ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ The SpecEnv of an Id maps a list of types (the template) to an expression [Type] |-> Expr For example, if f has this RuleInfo: [Int, a] -> \d:Ord Int. f' a it means that we can replace the call f Int t ===> (\d. f' t) This chucks one dictionary away and proceeds with the specialised version of f, namely f'. What can't be done this way? ~~~~~~~~~~~~~~~~~~~~~~~~~~~~ There is no way, post-typechecker, to get a dictionary for (say) Eq a from a dictionary for Eq [a]. So if we find ==.sel [t] d we can't transform to eqList (==.sel t d') where eqList :: (a->a->Bool) -> [a] -> [a] -> Bool Of course, we currently have no way to automatically derive eqList, nor to connect it to the Eq [a] instance decl, but you can imagine that it might somehow be possible. Taking advantage of this is permanently ruled out. Still, this is no great hardship, because we intend to eliminate overloading altogether anyway! A note about non-tyvar dictionaries ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Some Ids have types like forall a,b,c. Eq a -> Ord [a] -> tau This seems curious at first, because we usually only have dictionary args whose types are of the form (C a) where a is a type variable. But this doesn't hold for the functions arising from instance decls, which sometimes get arguments with types of form (C (T a)) for some type constructor T. Should we specialise wrt this compound-type dictionary? We used to say "no", saying: "This is a heuristic judgement, as indeed is the fact that we specialise wrt only dictionaries. We choose *not* to specialise wrt compound dictionaries because at the moment the only place they show up is in instance decls, where they are simply plugged into a returned dictionary. So nothing is gained by specialising wrt them." But it is simpler and more uniform to specialise wrt these dicts too; and in future GHC is likely to support full fledged type signatures like f :: Eq [(a,b)] => ... ************************************************************************ * * \subsubsection{The new specialiser} * * ************************************************************************ Our basic game plan is this. For let(rec) bound function f :: (C a, D c) => (a,b,c,d) -> Bool * Find any specialised calls of f, (f ts ds), where ts are the type arguments t1 .. t4, and ds are the dictionary arguments d1 .. d2. * Add a new definition for f1 (say): f1 = /\ b d -> (..body of f..) t1 b t3 d d1 d2 Note that we abstract over the unconstrained type arguments. * Add the mapping [t1,b,t3,d] |-> \d1 d2 -> f1 b d to the specialisations of f. This will be used by the simplifier to replace calls (f t1 t2 t3 t4) da db by (\d1 d1 -> f1 t2 t4) da db All the stuff about how many dictionaries to discard, and what types to apply the specialised function to, are handled by the fact that the SpecEnv contains a template for the result of the specialisation. We don't build *partial* specialisations for f. For example: f :: Eq a => a -> a -> Bool {-# SPECIALISE f :: (Eq b, Eq c) => (b,c) -> (b,c) -> Bool #-} Here, little is gained by making a specialised copy of f. There's a distinct danger that the specialised version would first build a dictionary for (Eq b, Eq c), and then select the (==) method from it! Even if it didn't, not a great deal is saved. We do, however, generate polymorphic, but not overloaded, specialisations: f :: Eq a => [a] -> b -> b -> b ... SPECIALISE f :: [Int] -> b -> b -> b ... Hence, the invariant is this: *** no specialised version is overloaded *** ************************************************************************ * * \subsubsection{The exported function} * * ************************************************************************ -} -- | Specialise calls to type-class overloaded functions occurring in a program. specProgram :: ModGuts -> CoreM ModGuts specProgram guts@(ModGuts { mg_module = this_mod , mg_rules = local_rules , mg_binds = binds }) = do { dflags <- getDynFlags -- We need to start with a Subst that knows all the things -- that are in scope, so that the substitution engine doesn't -- accidentally re-use a unique that's already in use -- Easiest thing is to do it all at once, as if all the top-level -- decls were mutually recursive ; let top_env = SE { se_subst = Core.mkEmptySubst $ mkInScopeSet $ mkVarSet $ bindersOfBinds binds , se_interesting = emptyVarSet , se_module = this_mod , se_dflags = dflags } go [] = return ([], emptyUDs) go (bind:binds) = do (binds', uds) <- go binds (bind', uds') <- specBind top_env bind uds return (bind' ++ binds', uds') -- Specialise the bindings of this module ; (binds', uds) <- runSpecM (go binds) ; (spec_rules, spec_binds) <- specImports top_env local_rules uds ; return (guts { mg_binds = spec_binds ++ binds' , mg_rules = spec_rules ++ local_rules }) } {- Note [Wrap bindings returned by specImports] ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 'specImports' returns a set of specialized bindings. However, these are lacking necessary floated dictionary bindings, which are returned by UsageDetails(ud_binds). These dictionaries need to be brought into scope with 'wrapDictBinds' before the bindings returned by 'specImports' can be used. See, for instance, the 'specImports' call in 'specProgram'. Note [Disabling cross-module specialisation] ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Since GHC 7.10 we have performed specialisation of INLINABLE bindings living in modules outside of the current module. This can sometimes uncover user code which explodes in size when aggressively optimized. The -fno-cross-module-specialise option was introduced to allow users to being bitten by such instances to revert to the pre-7.10 behavior. See #10491 -} {- ********************************************************************* * * Specialising imported functions * * ********************************************************************* -} specImports :: SpecEnv -> [CoreRule] -> UsageDetails -> CoreM ([CoreRule], [CoreBind]) specImports top_env local_rules (MkUD { ud_binds = dict_binds, ud_calls = calls }) | not $ gopt Opt_CrossModuleSpecialise (se_dflags top_env) -- See Note [Disabling cross-module specialisation] = return ([], wrapDictBinds dict_binds []) | otherwise = do { hpt_rules <- getRuleBase ; let rule_base = extendRuleBaseList hpt_rules local_rules ; (spec_rules, spec_binds) <- spec_imports top_env [] rule_base dict_binds calls -- Don't forget to wrap the specialized bindings with -- bindings for the needed dictionaries. -- See Note [Wrap bindings returned by specImports] -- and Note [Glom the bindings if imported functions are specialised] ; let final_binds | null spec_binds = wrapDictBinds dict_binds [] | otherwise = [Rec $ flattenBinds $ wrapDictBinds dict_binds spec_binds] ; return (spec_rules, final_binds) } -- | Specialise a set of calls to imported bindings spec_imports :: SpecEnv -- Passed in so that all top-level Ids are in scope -> [Id] -- Stack of imported functions being specialised -- See Note [specImport call stack] -> RuleBase -- Rules from this module and the home package -- (but not external packages, which can change) -> Bag DictBind -- Dict bindings, used /only/ for filterCalls -- See Note [Avoiding loops in specImports] -> CallDetails -- Calls for imported things -> CoreM ( [CoreRule] -- New rules , [CoreBind] ) -- Specialised bindings spec_imports top_env callers rule_base dict_binds calls = do { let import_calls = dVarEnvElts calls -- ; debugTraceMsg (text "specImports {" <+> -- vcat [ text "calls:" <+> ppr import_calls -- , text "dict_binds:" <+> ppr dict_binds ]) ; (rules, spec_binds) <- go rule_base import_calls -- ; debugTraceMsg (text "End specImports }" <+> ppr import_calls) ; return (rules, spec_binds) } where go :: RuleBase -> [CallInfoSet] -> CoreM ([CoreRule], [CoreBind]) go _ [] = return ([], []) go rb (cis : other_calls) = do { -- debugTraceMsg (text "specImport {" <+> ppr cis) ; (rules1, spec_binds1) <- spec_import top_env callers rb dict_binds cis -- ; debugTraceMsg (text "specImport }" <+> ppr cis) ; (rules2, spec_binds2) <- go (extendRuleBaseList rb rules1) other_calls ; return (rules1 ++ rules2, spec_binds1 ++ spec_binds2) } spec_import :: SpecEnv -- Passed in so that all top-level Ids are in scope -> [Id] -- Stack of imported functions being specialised -- See Note [specImport call stack] -> RuleBase -- Rules from this module -> Bag DictBind -- Dict bindings, used /only/ for filterCalls -- See Note [Avoiding loops in specImports] -> CallInfoSet -- Imported function and calls for it -> CoreM ( [CoreRule] -- New rules , [CoreBind] ) -- Specialised bindings spec_import top_env callers rb dict_binds cis@(CIS fn _) | isIn "specImport" fn callers = return ([], []) -- No warning. This actually happens all the time -- when specialising a recursive function, because -- the RHS of the specialised function contains a recursive -- call to the original function | null good_calls = return ([], []) | Just rhs <- canSpecImport dflags fn = do { -- Get rules from the external package state -- We keep doing this in case we "page-fault in" -- more rules as we go along ; hsc_env <- getHscEnv ; eps <- liftIO $ hscEPS hsc_env ; vis_orphs <- getVisibleOrphanMods ; let full_rb = unionRuleBase rb (eps_rule_base eps) rules_for_fn = getRules (RuleEnv full_rb vis_orphs) fn ; (rules1, spec_pairs, MkUD { ud_binds = dict_binds1, ud_calls = new_calls }) <- -- debugTraceMsg (text "specImport1" <+> vcat [ppr fn, ppr good_calls, ppr rhs]) >> (runSpecM $ specCalls True top_env rules_for_fn good_calls fn rhs) ; let spec_binds1 = [NonRec b r | (b,r) <- spec_pairs] -- After the rules kick in we may get recursion, but -- we rely on a global GlomBinds to sort that out later -- See Note [Glom the bindings if imported functions are specialised] -- Now specialise any cascaded calls -- ; debugTraceMsg (text "specImport 2" <+> (ppr fn $$ ppr rules1 $$ ppr spec_binds1)) ; (rules2, spec_binds2) <- spec_imports top_env (fn:callers) (extendRuleBaseList rb rules1) (dict_binds `unionBags` dict_binds1) new_calls ; let final_binds = wrapDictBinds dict_binds1 $ spec_binds2 ++ spec_binds1 ; return (rules2 ++ rules1, final_binds) } | otherwise = do { tryWarnMissingSpecs dflags callers fn good_calls ; return ([], [])} where dflags = se_dflags top_env good_calls = filterCalls cis dict_binds -- SUPER IMPORTANT! Drop calls that (directly or indirectly) refer to fn -- See Note [Avoiding loops in specImports] canSpecImport :: DynFlags -> Id -> Maybe CoreExpr -- See Note [Specialise imported INLINABLE things] canSpecImport dflags fn | CoreUnfolding { uf_src = src, uf_tmpl = rhs } <- unf , isStableSource src = Just rhs -- By default, specialise only imported things that have a stable -- unfolding; that is, have an INLINE or INLINABLE pragma -- Specialise even INLINE things; it hasn't inlined yet, -- so perhaps it never will. Moreover it may have calls -- inside it that we want to specialise -- CoreUnfolding case does /not/ include DFunUnfoldings; -- We only specialise DFunUnfoldings with -fspecialise-aggressively -- See Note [Do not specialise imported DFuns] | gopt Opt_SpecialiseAggressively dflags = maybeUnfoldingTemplate unf -- With -fspecialise-aggressively, specialise anything -- with an unfolding, stable or not, DFun or not | otherwise = Nothing where unf = realIdUnfolding fn -- We want to see the unfolding even for loop breakers -- | Returns whether or not to show a missed-spec warning. -- If -Wall-missed-specializations is on, show the warning. -- Otherwise, if -Wmissed-specializations is on, only show a warning -- if there is at least one imported function being specialized, -- and if all imported functions are marked with an inline pragma -- Use the most specific warning as the reason. tryWarnMissingSpecs :: DynFlags -> [Id] -> Id -> [CallInfo] -> CoreM () -- See Note [Warning about missed specialisations] tryWarnMissingSpecs dflags callers fn calls_for_fn | isClassOpId fn = return () -- See Note [Missed specialization for ClassOps] | wopt Opt_WarnMissedSpecs dflags && not (null callers) && allCallersInlined = doWarn $ Reason Opt_WarnMissedSpecs | wopt Opt_WarnAllMissedSpecs dflags = doWarn $ Reason Opt_WarnAllMissedSpecs | otherwise = return () where allCallersInlined = all (isAnyInlinePragma . idInlinePragma) callers doWarn reason = warnMsg reason (vcat [ hang (text ("Could not specialise imported function") <+> quotes (ppr fn)) 2 (vcat [ text "when specialising" <+> quotes (ppr caller) | caller <- callers]) , whenPprDebug (text "calls:" <+> vcat (map (pprCallInfo fn) calls_for_fn)) , text "Probable fix: add INLINABLE pragma on" <+> quotes (ppr fn) ]) {- Note [Missed specialisation for ClassOps] ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ In #19592 I saw a number of missed specialisation warnings which were the result of things like: case isJumpishInstr @X86.Instr $dInstruction_s7f8 eta3_a78C of { ... where isJumpishInstr is part of the Instruction class and defined like this: class Instruction instr where ... isJumpishInstr :: instr -> Bool ... isJumpishInstr is a ClassOp which will select the right method from within the dictionary via our built in rules. See also Note [ClassOp/DFun selection] in GHC.Tc.TyCl.Instance. We don't give these unfoldings, and as a result the specialiser complains. But usually this doesn't matter. The simplifier will apply the rule and we end up with case isJumpishInstrImplX86 eta3_a78C of { ... Since isJumpishInstrImplX86 is defined for a concrete instance (given by the dictionary) it is usually already well specialised! Theoretically the implementation of a method could still be overloaded over a different type class than what it's a method of. But I wasn't able to make this go wrong, and SPJ thinks this should be fine as well. So I decided to remove the warnings for failed specialisations on ClassOps alltogether as they do more harm than good. -} {- Note [Do not specialise imported DFuns] ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Ticket #18223 shows that specialising calls of DFuns is can cause a huge and entirely unnecessary blowup in program size. Consider a call to f @[[[[[[[[T]]]]]]]] d1 x where df :: C a => C [a] d1 :: C [[[[[[[[T]]]]]]]] = dfC[] @[[[[[[[T]]]]]]] d1 d2 :: C [[[[[[[T]]]]]]] = dfC[] @[[[[[[T]]]]]] d3 ... Now we'll specialise f's RHS, which may give rise to calls to 'g', also overloaded, which we will specialise, and so on. However, if we specialise the calls to dfC[], we'll generate specialised copies of all methods of C, at all types; and the same for C's superclasses. And many of these specialised functions will never be called. We are going to call the specialised 'f', and the specialised 'g', but DFuns group functions into a tuple, many of whose elements may never be used. With deeply-nested types this can lead to a simply overwhelming number of specialisations: see #18223 for a simple example (from the wild). I measured the number of specialisations for various numbers of calls of `flip evalStateT ()`, and got this Size after one simplification #calls #SPEC rules Terms Types 5 56 3100 10600 9 108 13660 77206 The real tests case has 60+ calls, which blew GHC out of the water. Solution: don't specialise DFuns. The downside is that if we end up with (h (dfun d)), /and/ we don't specialise 'h', then we won't pass to 'h' a tuple of specialised functions. However, the flag -fspecialise-aggressively (experimental, off by default) allows DFuns to specialise as well. Note [Avoiding loops in specImports] ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ We must take great care when specialising instance declarations (functions like $fOrdList) lest we accidentally build a recursive dictionary. See Note [Avoiding loops]. The basic strategy of Note [Avoiding loops] is to use filterCalls to discard loopy specialisations. But to do that we must ensure that the in-scope dict-binds (passed to filterCalls) contains all the needed dictionary bindings. In particular, in the recursive call to spec_imorpts in spec_import, we must include the dict-binds from the parent. Lacking this caused #17151, a really nasty bug. Here is what happened. * Class structure: Source is a superclass of Mut Index is a superclass of Source * We started with these dict binds dSource = $fSourcePix @Int $fIndexInt dIndex = sc_sel dSource dMut = $fMutPix @Int dIndex and these calls to specialise $fMutPix @Int dIndex $fSourcePix @Int $fIndexInt * We specialised the call ($fMutPix @Int dIndex) ==> new call ($fSourcePix @Int dIndex) (because Source is a superclass of Mut) * We specialised ($fSourcePix @Int dIndex) ==> produces specialised dict $s$fSourcePix, a record with dIndex as a field plus RULE forall d. ($fSourcePix @Int d) = $s$fSourcePix *** This is the bogus step *** * Now we decide not to specialise the call $fSourcePix @Int $fIndexInt because we alredy have a RULE that matches it * Finally the simplifer rewrites dSource = $fSourcePix @Int $fIndexInt ==> dSource = $s$fSourcePix Disaster. Now we have Rewrite dSource's RHS to $s$fSourcePix Disaster dSource = $s$fSourcePix dIndex = sc_sel dSource $s$fSourcePix = MkSource dIndex ... Solution: filterCalls should have stopped the bogus step, by seeing that dIndex transitively uses $fSourcePix. But it can only do that if it sees all the dict_binds. Wow. -------------- Here's another example (#13429). Suppose we have class Monoid v => C v a where ... We start with a call f @ [Integer] @ Integer $fC[]Integer Specialising call to 'f' gives dict bindings $dMonoid_1 :: Monoid [Integer] $dMonoid_1 = M.$p1C @ [Integer] $fC[]Integer $dC_1 :: C [Integer] (Node [Integer] Integer) $dC_1 = M.$fCvNode @ [Integer] $dMonoid_1 ...plus a recursive call to f @ [Integer] @ (Node [Integer] Integer) $dC_1 Specialising that call gives $dMonoid_2 :: Monoid [Integer] $dMonoid_2 = M.$p1C @ [Integer] $dC_1 $dC_2 :: C [Integer] (Node [Integer] Integer) $dC_2 = M.$fCvNode @ [Integer] $dMonoid_2 Now we have two calls to the imported function M.$fCvNode :: Monoid v => C v a M.$fCvNode @v @a m = C m some_fun But we must /not/ use the call (M.$fCvNode @ [Integer] $dMonoid_2) for specialisation, else we get: $dC_1 = M.$fCvNode @ [Integer] $dMonoid_1 $dMonoid_2 = M.$p1C @ [Integer] $dC_1 $s$fCvNode = C $dMonoid_2 ... RULE M.$fCvNode [Integer] _ _ = $s$fCvNode Now use the rule to rewrite the call in the RHS of $dC_1 and we get a loop! Note [specImport call stack] ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ When specialising an imports function 'f', we may get new calls of an imported function 'g', which we want to specialise in turn, and similarly specialising 'g' might expose a new call to 'h'. We track the stack of enclosing functions. So when specialising 'h' we haev a specImport call stack of [g,f]. We do this for two reasons: * Note [Warning about missed specialisations] * Note [Avoiding recursive specialisation] Note [Warning about missed specialisations] ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Suppose * In module Lib, you carefully mark a function 'foo' INLINABLE * Import Lib(foo) into another module M * Call 'foo' at some specialised type in M Then you jolly well expect it to be specialised in M. But what if 'foo' calls another function 'Lib.bar'. Then you'd like 'bar' to be specialised too. But if 'bar' is not marked INLINABLE it may well not be specialised. The warning Opt_WarnMissedSpecs warns about this. It's more noisy to warning about a missed specialisation opportunity for /every/ overloaded imported function, but sometimes useful. That is what Opt_WarnAllMissedSpecs does. ToDo: warn about missed opportunities for local functions. Note [Avoiding recursive specialisation] ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ When we specialise 'f' we may find new overloaded calls to 'g', 'h' in 'f's RHS. So we want to specialise g,h. But we don't want to specialise f any more! It's possible that f's RHS might have a recursive yet-more-specialised call, so we'd diverge in that case. And if the call is to the same type, one specialisation is enough. Avoiding this recursive specialisation loop is one reason for the 'callers' stack passed to specImports and specImport. Note [Specialise imported INLINABLE things] ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ What imported functions do we specialise? The basic set is * DFuns and things with INLINABLE pragmas. but with -fspecialise-aggressively we add * Anything with an unfolding template #8874 has a good example of why we want to auto-specialise DFuns. We have the -fspecialise-aggressively flag (usually off), because we risk lots of orphan modules from over-vigorous specialisation. However it's not a big deal: anything non-recursive with an unfolding-template will probably have been inlined already. Note [Glom the bindings if imported functions are specialised] ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Suppose we have an imported, *recursive*, INLINABLE function f :: Eq a => a -> a f = /\a \d x. ...(f a d)... In the module being compiled we have g x = f (x::Int) Now we'll make a specialised function f_spec :: Int -> Int f_spec = \x -> ...(f Int dInt)... {-# RULE f Int _ = f_spec #-} g = \x. f Int dInt x Note that f_spec doesn't look recursive After rewriting with the RULE, we get f_spec = \x -> ...(f_spec)... BUT since f_spec was non-recursive before it'll *stay* non-recursive. The occurrence analyser never turns a NonRec into a Rec. So we must make sure that f_spec is recursive. Easiest thing is to make all the specialisations for imported bindings recursive. ************************************************************************ * * \subsubsection{@specExpr@: the main function} * * ************************************************************************ -} data SpecEnv = SE { se_subst :: Core.Subst -- We carry a substitution down: -- a) we must clone any binding that might float outwards, -- to avoid name clashes -- b) we carry a type substitution to use when analysing -- the RHS of specialised bindings (no type-let!) , se_interesting :: VarSet -- Dict Ids that we know something about -- and hence may be worth specialising against -- See Note [Interesting dictionary arguments] , se_module :: Module , se_dflags :: DynFlags } instance Outputable SpecEnv where ppr (SE { se_subst = subst, se_interesting = interesting }) = text "SE" <+> braces (sep $ punctuate comma [ text "subst =" <+> ppr subst , text "interesting =" <+> ppr interesting ]) specVar :: SpecEnv -> Id -> CoreExpr specVar env v = Core.lookupIdSubst (se_subst env) v specExpr :: SpecEnv -> CoreExpr -> SpecM (CoreExpr, UsageDetails) ---------------- First the easy cases -------------------- specExpr env (Type ty) = return (Type (substTy env ty), emptyUDs) specExpr env (Coercion co) = return (Coercion (substCo env co), emptyUDs) specExpr env (Var v) = return (specVar env v, emptyUDs) specExpr _ (Lit lit) = return (Lit lit, emptyUDs) specExpr env (Cast e co) = do { (e', uds) <- specExpr env e ; return ((mkCast e' (substCo env co)), uds) } specExpr env (Tick tickish body) = do { (body', uds) <- specExpr env body ; return (Tick (specTickish env tickish) body', uds) } ---------------- Applications might generate a call instance -------------------- specExpr env expr@(App {}) = go expr [] where go (App fun arg) args = do (arg', uds_arg) <- specExpr env arg (fun', uds_app) <- go fun (arg':args) return (App fun' arg', uds_arg `plusUDs` uds_app) go (Var f) args = case specVar env f of Var f' -> return (Var f', mkCallUDs env f' args) e' -> return (e', emptyUDs) -- I don't expect this! go other _ = specExpr env other ---------------- Lambda/case require dumping of usage details -------------------- specExpr env e@(Lam {}) = specLam env' bndrs' body where (bndrs, body) = collectBinders e (env', bndrs') = substBndrs env bndrs -- More efficient to collect a group of binders together all at once -- and we don't want to split a lambda group with dumped bindings specExpr env (Case scrut case_bndr ty alts) = do { (scrut', scrut_uds) <- specExpr env scrut ; (scrut'', case_bndr', alts', alts_uds) <- specCase env scrut' case_bndr alts ; return (Case scrut'' case_bndr' (substTy env ty) alts' , scrut_uds `plusUDs` alts_uds) } ---------------- Finally, let is the interesting case -------------------- specExpr env (Let bind body) = do { -- Clone binders (rhs_env, body_env, bind') <- cloneBindSM env bind -- Deal with the body ; (body', body_uds) <- specExpr body_env body -- Deal with the bindings ; (binds', uds) <- specBind rhs_env bind' body_uds -- All done ; return (foldr Let body' binds', uds) } -------------- specLam :: SpecEnv -> [OutBndr] -> InExpr -> SpecM (OutExpr, UsageDetails) -- The binders have been substituted, but the body has not specLam env bndrs body | null bndrs = specExpr env body | otherwise = do { (body', uds) <- specExpr env body ; let (free_uds, dumped_dbs) = dumpUDs bndrs uds ; return (mkLams bndrs (wrapDictBindsE dumped_dbs body'), free_uds) } -------------- specTickish :: SpecEnv -> CoreTickish -> CoreTickish specTickish env (Breakpoint ext ix ids) = Breakpoint ext ix [ id' | id <- ids, Var id' <- [specVar env id]] -- drop vars from the list if they have a non-variable substitution. -- should never happen, but it's harmless to drop them anyway. specTickish _ other_tickish = other_tickish -------------- specCase :: SpecEnv -> CoreExpr -- Scrutinee, already done -> Id -> [CoreAlt] -> SpecM ( CoreExpr -- New scrutinee , Id , [CoreAlt] , UsageDetails) specCase env scrut' case_bndr [Alt con args rhs] | isDictId case_bndr -- See Note [Floating dictionaries out of cases] , interestingDict env scrut' , not (isDeadBinder case_bndr && null sc_args') = do { (case_bndr_flt : sc_args_flt) <- mapM clone_me (case_bndr' : sc_args') ; let sc_rhss = [ Case (Var case_bndr_flt) case_bndr' (idType sc_arg') [Alt con args' (Var sc_arg')] | sc_arg' <- sc_args' ] -- Extend the substitution for RHS to map the *original* binders -- to their floated versions. mb_sc_flts :: [Maybe DictId] mb_sc_flts = map (lookupVarEnv clone_env) args' clone_env = zipVarEnv sc_args' sc_args_flt subst_prs = (case_bndr, Var case_bndr_flt) : [ (arg, Var sc_flt) | (arg, Just sc_flt) <- args `zip` mb_sc_flts ] env_rhs' = env_rhs { se_subst = Core.extendIdSubstList (se_subst env_rhs) subst_prs , se_interesting = se_interesting env_rhs `extendVarSetList` (case_bndr_flt : sc_args_flt) } ; (rhs', rhs_uds) <- specExpr env_rhs' rhs ; let scrut_bind = mkDB (NonRec case_bndr_flt scrut') case_bndr_set = unitVarSet case_bndr_flt sc_binds = [ DB { db_bind = NonRec sc_arg_flt sc_rhs , db_fvs = case_bndr_set } | (sc_arg_flt, sc_rhs) <- sc_args_flt `zip` sc_rhss ] flt_binds = scrut_bind : sc_binds (free_uds, dumped_dbs) = dumpUDs (case_bndr':args') rhs_uds all_uds = flt_binds `addDictBinds` free_uds alt' = Alt con args' (wrapDictBindsE dumped_dbs rhs') ; return (Var case_bndr_flt, case_bndr', [alt'], all_uds) } where (env_rhs, (case_bndr':args')) = substBndrs env (case_bndr:args) sc_args' = filter is_flt_sc_arg args' clone_me bndr = do { uniq <- getUniqueM ; return (mkUserLocalOrCoVar occ uniq wght ty loc) } where name = idName bndr wght = idMult bndr ty = idType bndr occ = nameOccName name loc = getSrcSpan name arg_set = mkVarSet args' is_flt_sc_arg var = isId var && not (isDeadBinder var) && isDictTy var_ty && tyCoVarsOfType var_ty `disjointVarSet` arg_set where var_ty = idType var specCase env scrut case_bndr alts = do { (alts', uds_alts) <- mapAndCombineSM spec_alt alts ; return (scrut, case_bndr', alts', uds_alts) } where (env_alt, case_bndr') = substBndr env case_bndr spec_alt (Alt con args rhs) = do (rhs', uds) <- specExpr env_rhs rhs let (free_uds, dumped_dbs) = dumpUDs (case_bndr' : args') uds return (Alt con args' (wrapDictBindsE dumped_dbs rhs'), free_uds) where (env_rhs, args') = substBndrs env_alt args {- Note [Floating dictionaries out of cases] ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Consider g = \d. case d of { MkD sc ... -> ...(f sc)... } Naively we can't float d2's binding out of the case expression, because 'sc' is bound by the case, and that in turn means we can't specialise f, which seems a pity. So we invert the case, by floating out a binding for 'sc_flt' thus: sc_flt = case d of { MkD sc ... -> sc } Now we can float the call instance for 'f'. Indeed this is just what'll happen if 'sc' was originally bound with a let binding, but case is more efficient, and necessary with equalities. So it's good to work with both. You might think that this won't make any difference, because the call instance will only get nuked by the \d. BUT if 'g' itself is specialised, then transitively we should be able to specialise f. In general, given case e of cb { MkD sc ... -> ...(f sc)... } we transform to let cb_flt = e sc_flt = case cb_flt of { MkD sc ... -> sc } in case cb_flt of bg { MkD sc ... -> ....(f sc_flt)... } The "_flt" things are the floated binds; we use the current substitution to substitute sc -> sc_flt in the RHS ************************************************************************ * * Dealing with a binding * * ************************************************************************ -} specBind :: SpecEnv -- Use this for RHSs -> CoreBind -- Binders are already cloned by cloneBindSM, -- but RHSs are un-processed -> UsageDetails -- Info on how the scope of the binding -> SpecM ([CoreBind], -- New bindings UsageDetails) -- And info to pass upstream -- Returned UsageDetails: -- No calls for binders of this bind specBind rhs_env (NonRec fn rhs) body_uds = do { (rhs', rhs_uds) <- specExpr rhs_env rhs ; let zapped_fn = zapIdDemandInfo fn -- We zap the demand info because the binding may float, -- which would invaidate the demand info (see #17810 for example). -- Destroying demand info is not terrible; specialisation is -- always followed soon by demand analysis. ; (fn', spec_defns, body_uds1) <- specDefn rhs_env body_uds zapped_fn rhs ; let pairs = spec_defns ++ [(fn', rhs')] -- fn' mentions the spec_defns in its rules, -- so put the latter first combined_uds = body_uds1 `plusUDs` rhs_uds (free_uds, dump_dbs, float_all) = dumpBindUDs [fn] combined_uds final_binds :: [DictBind] -- See Note [From non-recursive to recursive] final_binds | not (isEmptyBag dump_dbs) , not (null spec_defns) = [recWithDumpedDicts pairs dump_dbs] | otherwise = [mkDB $ NonRec b r | (b,r) <- pairs] ++ bagToList dump_dbs ; if float_all then -- Rather than discard the calls mentioning the bound variables -- we float this (dictionary) binding along with the others return ([], free_uds `snocDictBinds` final_binds) else -- No call in final_uds mentions bound variables, -- so we can just leave the binding here return (map db_bind final_binds, free_uds) } specBind rhs_env (Rec pairs) body_uds -- Note [Specialising a recursive group] = do { let (bndrs,rhss) = unzip pairs ; (rhss', rhs_uds) <- mapAndCombineSM (specExpr rhs_env) rhss ; let scope_uds = body_uds `plusUDs` rhs_uds -- Includes binds and calls arising from rhss ; (bndrs1, spec_defns1, uds1) <- specDefns rhs_env scope_uds pairs ; (bndrs3, spec_defns3, uds3) <- if null spec_defns1 -- Common case: no specialisation then return (bndrs1, [], uds1) else do { -- Specialisation occurred; do it again (bndrs2, spec_defns2, uds2) <- specDefns rhs_env uds1 (bndrs1 `zip` rhss) ; return (bndrs2, spec_defns2 ++ spec_defns1, uds2) } ; let (final_uds, dumped_dbs, float_all) = dumpBindUDs bndrs uds3 final_bind = recWithDumpedDicts (spec_defns3 ++ zip bndrs3 rhss') dumped_dbs ; if float_all then return ([], final_uds `snocDictBind` final_bind) else return ([db_bind final_bind], final_uds) } --------------------------- specDefns :: SpecEnv -> UsageDetails -- Info on how it is used in its scope -> [(OutId,InExpr)] -- The things being bound and their un-processed RHS -> SpecM ([OutId], -- Original Ids with RULES added [(OutId,OutExpr)], -- Extra, specialised bindings UsageDetails) -- Stuff to fling upwards from the specialised versions -- Specialise a list of bindings (the contents of a Rec), but flowing usages -- upwards binding by binding. Example: { f = ...g ...; g = ...f .... } -- Then if the input CallDetails has a specialised call for 'g', whose specialisation -- in turn generates a specialised call for 'f', we catch that in this one sweep. -- But not vice versa (it's a fixpoint problem). specDefns _env uds [] = return ([], [], uds) specDefns env uds ((bndr,rhs):pairs) = do { (bndrs1, spec_defns1, uds1) <- specDefns env uds pairs ; (bndr1, spec_defns2, uds2) <- specDefn env uds1 bndr rhs ; return (bndr1 : bndrs1, spec_defns1 ++ spec_defns2, uds2) } --------------------------- specDefn :: SpecEnv -> UsageDetails -- Info on how it is used in its scope -> OutId -> InExpr -- The thing being bound and its un-processed RHS -> SpecM (Id, -- Original Id with added RULES [(Id,CoreExpr)], -- Extra, specialised bindings UsageDetails) -- Stuff to fling upwards from the specialised versions specDefn env body_uds fn rhs = do { let (body_uds_without_me, calls_for_me) = callsForMe fn body_uds rules_for_me = idCoreRules fn ; (rules, spec_defns, spec_uds) <- specCalls False env rules_for_me calls_for_me fn rhs ; return ( fn `addIdSpecialisations` rules , spec_defns , body_uds_without_me `plusUDs` spec_uds) } -- It's important that the `plusUDs` is this way -- round, because body_uds_without_me may bind -- dictionaries that are used in calls_for_me passed -- to specDefn. So the dictionary bindings in -- spec_uds may mention dictionaries bound in -- body_uds_without_me --------------------------- specCalls :: Bool -- True => specialising imported fn -- False => specialising local fn -> SpecEnv -> [CoreRule] -- Existing RULES for the fn -> [CallInfo] -> OutId -> InExpr -> SpecM SpecInfo -- New rules, specialised bindings, and usage details -- This function checks existing rules, and does not create -- duplicate ones. So the caller does not need to do this filtering. -- See 'already_covered' type SpecInfo = ( [CoreRule] -- Specialisation rules , [(Id,CoreExpr)] -- Specialised definition , UsageDetails ) -- Usage details from specialised RHSs specCalls spec_imp env existing_rules calls_for_me fn rhs -- The first case is the interesting one | notNull calls_for_me -- And there are some calls to specialise && not (isNeverActive (idInlineActivation fn)) -- Don't specialise NOINLINE things -- See Note [Auto-specialisation and RULES] -- && not (certainlyWillInline (idUnfolding fn)) -- And it's not small -- See Note [Inline specialisation] for why we do not -- switch off specialisation for inline functions = -- pprTrace "specDefn: some" (ppr fn $$ ppr calls_for_me $$ ppr existing_rules) $ foldlM spec_call ([], [], emptyUDs) calls_for_me | otherwise -- No calls or RHS doesn't fit our preconceptions = WARN( not (exprIsTrivial rhs) && notNull calls_for_me, text "Missed specialisation opportunity for" <+> ppr fn $$ _trace_doc ) -- Note [Specialisation shape] -- pprTrace "specDefn: none" (ppr fn <+> ppr calls_for_me) $ return ([], [], emptyUDs) where _trace_doc = sep [ ppr rhs_bndrs, ppr (idInlineActivation fn) ] fn_type = idType fn fn_arity = idArity fn fn_unf = realIdUnfolding fn -- Ignore loop-breaker-ness here inl_prag = idInlinePragma fn inl_act = inlinePragmaActivation inl_prag is_local = isLocalId fn is_dfun = isDFunId fn dflags = se_dflags env ropts = initRuleOpts dflags this_mod = se_module env -- Figure out whether the function has an INLINE pragma -- See Note [Inline specialisations] (rhs_bndrs, rhs_body) = collectBindersPushingCo rhs -- See Note [Account for casts in binding] in_scope = Core.substInScope (se_subst env) already_covered :: RuleOpts -> [CoreRule] -> [CoreExpr] -> Bool already_covered ropts new_rules args -- Note [Specialisations already covered] = isJust (lookupRule ropts (in_scope, realIdUnfolding) (const True) fn args (new_rules ++ existing_rules)) -- NB: we look both in the new_rules (generated by this invocation -- of specCalls), and in existing_rules (passed in to specCalls) ---------------------------------------------------------- -- Specialise to one particular call pattern spec_call :: SpecInfo -- Accumulating parameter -> CallInfo -- Call instance -> SpecM SpecInfo spec_call spec_acc@(rules_acc, pairs_acc, uds_acc) _ci@(CI { ci_key = call_args }) = -- See Note [Specialising Calls] do { let all_call_args | is_dfun = call_args ++ repeat UnspecArg | otherwise = call_args -- See Note [Specialising DFuns] ; ( useful, rhs_env2, leftover_bndrs , rule_bndrs, rule_lhs_args , spec_bndrs1, dx_binds, spec_args) <- specHeader env rhs_bndrs all_call_args -- ; pprTrace "spec_call" (vcat [ text "call info: " <+> ppr _ci -- , text "useful: " <+> ppr useful -- , text "rule_bndrs:" <+> ppr rule_bndrs -- , text "lhs_args: " <+> ppr rule_lhs_args -- , text "spec_bndrs:" <+> ppr spec_bndrs1 -- , text "spec_args: " <+> ppr spec_args -- , text "dx_binds: " <+> ppr dx_binds -- , text "rhs_env2: " <+> ppr (se_subst rhs_env2) -- , ppr dx_binds ]) $ -- return () ; if not useful -- No useful specialisation || already_covered ropts rules_acc rule_lhs_args then return spec_acc else do { -- Run the specialiser on the specialised RHS -- The "1" suffix is before we maybe add the void arg ; (spec_rhs1, rhs_uds) <- specLam rhs_env2 (spec_bndrs1 ++ leftover_bndrs) rhs_body ; let spec_fn_ty1 = exprType spec_rhs1 -- Maybe add a void arg to the specialised function, -- to avoid unlifted bindings -- See Note [Specialisations Must Be Lifted] -- C.f. GHC.Core.Opt.WorkWrap.Utils.mkWorkerArgs add_void_arg = isUnliftedType spec_fn_ty1 && not (isJoinId fn) (spec_bndrs, spec_rhs, spec_fn_ty) | add_void_arg = ( voidPrimId : spec_bndrs1 , Lam voidArgId spec_rhs1 , mkVisFunTyMany unboxedUnitTy spec_fn_ty1) | otherwise = (spec_bndrs1, spec_rhs1, spec_fn_ty1) join_arity_decr = length rule_lhs_args - length spec_bndrs spec_join_arity | Just orig_join_arity <- isJoinId_maybe fn = Just (orig_join_arity - join_arity_decr) | otherwise = Nothing ; spec_fn <- newSpecIdSM fn spec_fn_ty spec_join_arity ; let -- The rule to put in the function's specialisation is: -- forall x @b d1' d2'. -- f x @T1 @b @T2 d1' d2' = f1 x @b -- See Note [Specialising Calls] herald | spec_imp = -- Specialising imported fn text "SPEC/" <> ppr this_mod | otherwise = -- Specialising local fn text "SPEC" rule_name = mkFastString $ showSDoc dflags $ herald <+> ftext (occNameFS (getOccName fn)) <+> hsep (mapMaybe ppr_call_key_ty call_args) -- This name ends up in interface files, so use occNameString. -- Otherwise uniques end up there, making builds -- less deterministic (See #4012 comment:61 ff) rule_wout_eta = mkRule this_mod True {- Auto generated -} is_local rule_name inl_act -- Note [Auto-specialisation and RULES] (idName fn) rule_bndrs rule_lhs_args (mkVarApps (Var spec_fn) spec_bndrs) spec_rule = case isJoinId_maybe fn of Just join_arity -> etaExpandToJoinPointRule join_arity rule_wout_eta Nothing -> rule_wout_eta -- Add the { d1' = dx1; d2' = dx2 } usage stuff -- See Note [Specialising Calls] spec_uds = foldr consDictBind rhs_uds dx_binds simpl_opts = initSimpleOpts dflags -------------------------------------- -- Add a suitable unfolding if the spec_inl_prag says so -- See Note [Inline specialisations] (spec_inl_prag, spec_unf) | not is_local && isStrongLoopBreaker (idOccInfo fn) = (neverInlinePragma, noUnfolding) -- See Note [Specialising imported functions] in "GHC.Core.Opt.OccurAnal" | InlinePragma { inl_inline = Inlinable } <- inl_prag = (inl_prag { inl_inline = NoUserInlinePrag }, noUnfolding) | otherwise = (inl_prag, specUnfolding simpl_opts spec_bndrs (`mkApps` spec_args) rule_lhs_args fn_unf) -------------------------------------- -- Adding arity information just propagates it a bit faster -- See Note [Arity decrease] in GHC.Core.Opt.Simplify -- Copy InlinePragma information from the parent Id. -- So if f has INLINE[1] so does spec_fn arity_decr = count isValArg rule_lhs_args - count isId spec_bndrs spec_f_w_arity = spec_fn `setIdArity` max 0 (fn_arity - arity_decr) `setInlinePragma` spec_inl_prag `setIdUnfolding` spec_unf `asJoinId_maybe` spec_join_arity _rule_trace_doc = vcat [ ppr fn <+> dcolon <+> ppr fn_type , ppr spec_fn <+> dcolon <+> ppr spec_fn_ty , ppr rhs_bndrs, ppr call_args , ppr spec_rule ] ; -- pprTrace "spec_call: rule" _rule_trace_doc return ( spec_rule : rules_acc , (spec_f_w_arity, spec_rhs) : pairs_acc , spec_uds `plusUDs` uds_acc ) } } {- Note [Specialising DFuns] ~~~~~~~~~~~~~~~~~~~~~~~~~~~~ DFuns have a special sort of unfolding (DFunUnfolding), and these are hard to specialise a DFunUnfolding to give another DFunUnfolding unless the DFun is fully applied (#18120). So, in the case of DFunIds we simply extend the CallKey with trailing UnspecArgs, so we'll generate a rule that completely saturates the DFun. There is an ASSERT that checks this, in the DFunUnfolding case of GHC.Core.Unfold.specUnfolding. Note [Specialisation Must Preserve Sharing] ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Consider a function: f :: forall a. Eq a => a -> blah f = if expensive then f1 else f2 As written, all calls to 'f' will share 'expensive'. But if we specialise 'f' at 'Int', eg: $sfInt = SUBST[a->Int,dict->dEqInt] (if expensive then f1 else f2) RULE "SPEC f" forall (d :: Eq Int). f Int _ = $sfIntf We've now lost sharing between 'f' and '$sfInt' for 'expensive'. Yikes! To avoid this, we only generate specialisations for functions whose arity is enough to bind all of the arguments we need to specialise. This ensures our specialised functions don't do any work before receiving all of their dicts, and thus avoids the 'f' case above. Note [Specialisations Must Be Lifted] ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Consider a function 'f': f = forall a. Eq a => Array# a used like case x of True -> ...f @Int dEqInt... False -> 0 Naively, we might generate an (expensive) specialisation $sfInt :: Array# Int even in the case that @x = False@! Instead, we add a dummy 'Void#' argument to the specialisation '$sfInt' ($sfInt :: Void# -> Array# Int) in order to preserve laziness. Note [Specialising Calls] ~~~~~~~~~~~~~~~~~~~~~~~~~ Suppose we have a function with a complicated type: f :: forall a b c. Int -> Eq a => Show b => c -> Blah f @a @b @c i dEqA dShowA x = blah and suppose it is called at: f 7 @T1 @T2 @T3 dEqT1 ($dfShow dShowT2) t3 This call is described as a 'CallInfo' whose 'ci_key' is: [ SpecType T1, SpecType T2, UnspecType, UnspecArg, SpecDict dEqT1 , SpecDict ($dfShow dShowT2), UnspecArg ] Why are 'a' and 'b' identified as 'SpecType', while 'c' is 'UnspecType'? Because we must specialise the function on type variables that appear free in its *dictionary* arguments; but not on type variables that do not appear in any dictionaries, i.e. are fully polymorphic. Because this call has dictionaries applied, we'd like to specialise the call on any type argument that appears free in those dictionaries. In this case, those are [a :-> T1, b :-> T2]. We also need to substitute the dictionary binders with their specialised dictionaries. The simplest substitution would be [dEqA :-> dEqT1, dShowA :-> $dfShow dShowT2], but this duplicates work, since `$dfShow dShowT2` is a function application. Therefore, we also want to *float the dictionary out* (via bindAuxiliaryDict), creating a new dict binding dShow1 = $dfShow dShowT2 and the substitution [dEqA :-> dEqT1, dShowA :-> dShow1]. With the substitutions in hand, we can generate a specialised function: $sf :: forall c. Int -> c -> Blah $sf = SUBST[a :-> T1, b :-> T2, dEqA :-> dEqT1, dShowA :-> dShow1] (\@c i x -> blah) Note that the substitution is applied to the whole thing. This is convenient, but just slightly fragile. Notably: * There had better be no name clashes in a/b/c We must construct a rewrite rule: RULE "SPEC f @T1 @T2 _" forall (@c :: Type) (i :: Int) (d1 :: Eq T1) (d2 :: Show T2). f @T1 @T2 @c i d1 d2 = $sf @c i In the rule, d1 and d2 are just wildcards, not used in the RHS. Note additionally that 'x' isn't captured by this rule --- we bind only enough etas in order to capture all of the *specialised* arguments. Note [Drop dead args from specialisations] ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ When specialising a function, it’s possible some of the arguments may actually be dead. For example, consider: f :: forall a. () -> Show a => a -> String f x y = show y ++ "!" We might generate the following CallInfo for `f @Int`: [SpecType Int, UnspecArg, SpecDict $dShowInt, UnspecArg] Normally we’d include both the x and y arguments in the specialisation, since we’re not specialising on either of them. But that’s silly, since x is actually unused! So we might as well drop it in the specialisation: $sf :: Int -> String $sf y = show y ++ "!" {-# RULE "SPEC f @Int" forall x. f @Int x $dShow = $sf #-} This doesn’t save us much, since the arg would be removed later by worker/wrapper, anyway, but it’s easy to do. Note, however, that we only drop dead arguments if: 1. We don’t specialise on them. 2. They come before an argument we do specialise on. Doing the latter would require eta-expanding the RULE, which could make it match less often, so it’s not worth it. Doing the former could be more useful --- it would stop us from generating pointless specialisations --- but it’s more involved to implement and unclear if it actually provides much benefit in practice. Note [Zap occ info in rule binders] ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ When we generate a specialisation RULE, we need to drop occurrence info on the binders. If we don’t, things go wrong when we specialise a function like f :: forall a. () -> Show a => a -> String f x y = show y ++ "!" since we’ll generate a RULE like RULE "SPEC f @Int" forall x [Occ=Dead]. f @Int x $dShow = $sf and Core Lint complains, even though x only appears on the LHS (due to Note [Drop dead args from specialisations]). Why is that a Lint error? Because the arguments on the LHS of a rule are syntactically expressions, not patterns, so Lint treats the appearance of x as a use rather than a binding. Fortunately, the solution is simple: we just make sure to zap the occ info before using ids as wildcard binders in a rule. Note [Account for casts in binding] ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Consider f :: Eq a => a -> IO () {-# INLINABLE f StableUnf = (/\a \(d:Eq a) (x:a). blah) |> g #-} f = ... In f's stable unfolding we have done some modest simplification which has pushed the cast to the outside. (I wonder if this is the Right Thing, but it's what happens now; see GHC.Core.Opt.Simplify.Utils Note [Casts and lambdas].) Now that stable unfolding must be specialised, so we want to push the cast back inside. It would be terrible if the cast defeated specialisation! Hence the use of collectBindersPushingCo. Note [Evidence foralls] ~~~~~~~~~~~~~~~~~~~~~~~~~~ Suppose (#12212) that we are specialising f :: forall a b. (Num a, F a ~ F b) => blah with a=b=Int. Then the RULE will be something like RULE forall (d:Num Int) (g :: F Int ~ F Int). f Int Int d g = f_spec But both varToCoreExpr (when constructing the LHS args), and the simplifier (when simplifying the LHS args), will transform to RULE forall (d:Num Int) (g :: F Int ~ F Int). f Int Int d <F Int> = f_spec by replacing g with Refl. So now 'g' is unbound, which results in a later crash. So we use Refl right off the bat, and do not forall-quantify 'g': * varToCoreExpr generates a Refl * exprsFreeIdsList returns the Ids bound by the args, which won't include g You might wonder if this will match as often, but the simplifier replaces complicated Refl coercions with Refl pretty aggressively. Note [Orphans and auto-generated rules] ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ When we specialise an INLINABLE function, or when we have -fspecialise-aggressively, we auto-generate RULES that are orphans. We don't want to warn about these, or we'd generate a lot of warnings. Thus, we only warn about user-specified orphan rules. Indeed, we don't even treat the module as an orphan module if it has auto-generated *rule* orphans. Orphan modules are read every time we compile, so they are pretty obtrusive and slow down every compilation, even non-optimised ones. (Reason: for type class instances it's a type correctness issue.) But specialisation rules are strictly for *optimisation* only so it's fine not to read the interface. What this means is that a SPEC rules from auto-specialisation in module M will be used in other modules only if M.hi has been read for some other reason, which is actually pretty likely. Note [From non-recursive to recursive] ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Even in the non-recursive case, if any dict-binds depend on 'fn' we might have built a recursive knot f a d x = <blah> MkUD { ud_binds = NonRec d7 (MkD ..f..) , ud_calls = ...(f T d7)... } The we generate Rec { fs x = <blah>[T/a, d7/d] f a d x = <blah> RULE f T _ = fs d7 = ...f... } Here the recursion is only through the RULE. However we definitely should /not/ make the Rec in this wildly common case: d = ... MkUD { ud_binds = NonRec d7 (...d...) , ud_calls = ...(f T d7)... } Here we want simply to add d to the floats, giving MkUD { ud_binds = NonRec d (...) NonRec d7 (...d...) , ud_calls = ...(f T d7)... } In general, we need only make this Rec if - there are some specialisations (spec_binds non-empty) - there are some dict_binds that depend on f (dump_dbs non-empty) Note [Avoiding loops] ~~~~~~~~~~~~~~~~~~~~~ When specialising /dictionary functions/ we must be very careful to avoid building loops. Here is an example that bit us badly, on several distinct occasions. Here is one: #3591 class Eq a => C a instance Eq [a] => C [a] This translates to dfun :: Eq [a] -> C [a] dfun a d = MkD a d (meth d) d4 :: Eq [T] = <blah> d2 :: C [T] = dfun T d4 d1 :: Eq [T] = $p1 d2 d3 :: C [T] = dfun T d1 None of these definitions is recursive. What happened was that we generated a specialisation: RULE forall d. dfun T d = dT :: C [T] dT = (MkD a d (meth d)) [T/a, d1/d] = MkD T d1 (meth d1) But now we use the RULE on the RHS of d2, to get d2 = dT = MkD d1 (meth d1) d1 = $p1 d2 and now d1 is bottom! The problem is that when specialising 'dfun' we should first dump "below" the binding all floated dictionary bindings that mention 'dfun' itself. So d2 and d3 (and hence d1) must be placed below 'dfun', and thus unavailable to it when specialising 'dfun'. That in turn means that the call (dfun T d1) must be discarded. On the other hand, the call (dfun T d4) is fine, assuming d4 doesn't mention dfun. Solution: Discard all calls that mention dictionaries that depend (directly or indirectly) on the dfun we are specialising. This is done by 'filterCalls' -------------- Here's yet another example class C a where { foo,bar :: [a] -> [a] } instance C Int where foo x = r_bar x bar xs = reverse xs r_bar :: C a => [a] -> [a] r_bar xs = bar (xs ++ xs) That translates to: r_bar a (c::C a) (xs::[a]) = bar a d (xs ++ xs) Rec { $fCInt :: C Int = MkC foo_help reverse foo_help (xs::[Int]) = r_bar Int $fCInt xs } The call (r_bar $fCInt) mentions $fCInt, which mentions foo_help, which mentions r_bar But we DO want to specialise r_bar at Int: Rec { $fCInt :: C Int = MkC foo_help reverse foo_help (xs::[Int]) = r_bar Int $fCInt xs r_bar a (c::C a) (xs::[a]) = bar a d (xs ++ xs) RULE r_bar Int _ = r_bar_Int r_bar_Int xs = bar Int $fCInt (xs ++ xs) } Note that, because of its RULE, r_bar joins the recursive group. (In this case it'll unravel a short moment later.) Note [Specialising a recursive group] ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Consider let rec { f x = ...g x'... ; g y = ...f y'.... } in f 'a' Here we specialise 'f' at Char; but that is very likely to lead to a specialisation of 'g' at Char. We must do the latter, else the whole point of specialisation is lost. But we do not want to keep iterating to a fixpoint, because in the presence of polymorphic recursion we might generate an infinite number of specialisations. So we use the following heuristic: * Arrange the rec block in dependency order, so far as possible (the occurrence analyser already does this) * Specialise it much like a sequence of lets * Then go through the block a second time, feeding call-info from the RHSs back in the bottom, as it were In effect, the ordering maxmimises the effectiveness of each sweep, and we do just two sweeps. This should catch almost every case of monomorphic recursion -- the exception could be a very knotted-up recursion with multiple cycles tied up together. This plan is implemented in the Rec case of specBindItself. Note [Specialisations already covered] ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ We obviously don't want to generate two specialisations for the same argument pattern. There are two wrinkles 1. We do the already-covered test in specDefn, not when we generate the CallInfo in mkCallUDs. We used to test in the latter place, but we now iterate the specialiser somewhat, and the Id at the call site might therefore not have all the RULES that we can see in specDefn 2. What about two specialisations where the second is an *instance* of the first? If the more specific one shows up first, we'll generate specialisations for both. If the *less* specific one shows up first, we *don't* currently generate a specialisation for the more specific one. (See the call to lookupRule in already_covered.) Reasons: (a) lookupRule doesn't say which matches are exact (bad reason) (b) if the earlier specialisation is user-provided, it's far from clear that we should auto-specialise further Note [Auto-specialisation and RULES] ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Consider: g :: Num a => a -> a g = ... f :: (Int -> Int) -> Int f w = ... {-# RULE f g = 0 #-} Suppose that auto-specialisation makes a specialised version of g::Int->Int That version won't appear in the LHS of the RULE for f. So if the specialisation rule fires too early, the rule for f may never fire. It might be possible to add new rules, to "complete" the rewrite system. Thus when adding RULE forall d. g Int d = g_spec also add RULE f g_spec = 0 But that's a bit complicated. For now we ask the programmer's help, by *copying the INLINE activation pragma* to the auto-specialised rule. So if g says {-# NOINLINE[2] g #-}, then the auto-spec rule will also not be active until phase 2. And that's what programmers should jolly well do anyway, even aside from specialisation, to ensure that g doesn't inline too early. This in turn means that the RULE would never fire for a NOINLINE thing so not much point in generating a specialisation at all. Note [Specialisation shape] ~~~~~~~~~~~~~~~~~~~~~~~~~~~ We only specialise a function if it has visible top-level lambdas corresponding to its overloading. E.g. if f :: forall a. Eq a => .... then its body must look like f = /\a. \d. ... Reason: when specialising the body for a call (f ty dexp), we want to substitute dexp for d, and pick up specialised calls in the body of f. We do allow casts, however; see Note [Account for casts in binding]. This doesn't always work. One example I came across was this: newtype Gen a = MkGen{ unGen :: Int -> a } choose :: Eq a => a -> Gen a choose n = MkGen (\r -> n) oneof = choose (1::Int) It's a silly example, but we get choose = /\a. g `cast` co where choose doesn't have any dict arguments. Thus far I have not tried to fix this (wait till there's a real example). Mind you, then 'choose' will be inlined (since RHS is trivial) so it doesn't matter. This comes up with single-method classes class C a where { op :: a -> a } instance C a => C [a] where .... ==> $fCList :: C a => C [a] $fCList = $copList |> (...coercion>...) ....(uses of $fCList at particular types)... So we suppress the WARN if the rhs is trivial. Note [Inline specialisations] ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Here is what we do with the InlinePragma of the original function * Activation/RuleMatchInfo: both transferred to the specialised function * InlineSpec: (a) An INLINE pragma is transferred (b) An INLINABLE pragma is *not* transferred Why (a): transfer INLINE pragmas? The point of INLINE was precisely to specialise the function at its call site, and arguably that's not so important for the specialised copies. BUT *pragma-directed* specialisation now takes place in the typechecker/desugarer, with manually specified INLINEs. The specialisation here is automatic. It'd be very odd if a function marked INLINE was specialised (because of some local use), and then forever after (including importing modules) the specialised version wasn't INLINEd. After all, the programmer said INLINE! You might wonder why we specialise INLINE functions at all. After all they should be inlined, right? Two reasons: * Even INLINE functions are sometimes not inlined, when they aren't applied to interesting arguments. But perhaps the type arguments alone are enough to specialise (even though the args are too boring to trigger inlining), and it's certainly better to call the specialised version. * The RHS of an INLINE function might call another overloaded function, and we'd like to generate a specialised version of that function too. This actually happens a lot. Consider replicateM_ :: (Monad m) => Int -> m a -> m () {-# INLINABLE replicateM_ #-} replicateM_ d x ma = ... The strictness analyser may transform to replicateM_ :: (Monad m) => Int -> m a -> m () {-# INLINE replicateM_ #-} replicateM_ d x ma = case x of I# x' -> $wreplicateM_ d x' ma $wreplicateM_ :: (Monad m) => Int# -> m a -> m () {-# INLINABLE $wreplicateM_ #-} $wreplicateM_ = ... Now an importing module has a specialised call to replicateM_, say (replicateM_ dMonadIO). We certainly want to specialise $wreplicateM_! This particular example had a huge effect on the call to replicateM_ in nofib/shootout/n-body. Why (b): discard INLINABLE pragmas? See #4874 for persuasive examples. Suppose we have {-# INLINABLE f #-} f :: Ord a => [a] -> Int f xs = letrec f' = ...f'... in f' Then, when f is specialised and optimised we might get wgo :: [Int] -> Int# wgo = ...wgo... f_spec :: [Int] -> Int f_spec xs = case wgo xs of { r -> I# r } and we clearly want to inline f_spec at call sites. But if we still have the big, un-optimised of f (albeit specialised) captured in an INLINABLE pragma for f_spec, we won't get that optimisation. So we simply drop INLINABLE pragmas when specialising. It's not really a complete solution; ignoring specialisation for now, INLINABLE functions don't get properly strictness analysed, for example. But it works well for examples involving specialisation, which is the dominant use of INLINABLE. See #4874. -} {- ********************************************************************* * * SpecArg, and specHeader * * ********************************************************************* -} -- | An argument that we might want to specialise. -- See Note [Specialising Calls] for the nitty gritty details. data SpecArg = -- | Type arguments that should be specialised, due to appearing -- free in the type of a 'SpecDict'. SpecType Type -- | Type arguments that should remain polymorphic. | UnspecType -- | Dictionaries that should be specialised. mkCallUDs ensures -- that only "interesting" dictionary arguments get a SpecDict; -- see Note [Interesting dictionary arguments] | SpecDict DictExpr -- | Value arguments that should not be specialised. | UnspecArg instance Outputable SpecArg where ppr (SpecType t) = text "SpecType" <+> ppr t ppr UnspecType = text "UnspecType" ppr (SpecDict d) = text "SpecDict" <+> ppr d ppr UnspecArg = text "UnspecArg" specArgFreeVars :: SpecArg -> VarSet specArgFreeVars (SpecType ty) = tyCoVarsOfType ty specArgFreeVars (SpecDict dx) = exprFreeVars dx specArgFreeVars UnspecType = emptyVarSet specArgFreeVars UnspecArg = emptyVarSet isSpecDict :: SpecArg -> Bool isSpecDict (SpecDict {}) = True isSpecDict _ = False -- | Given binders from an original function 'f', and the 'SpecArg's -- corresponding to its usage, compute everything necessary to build -- a specialisation. -- -- We will use the running example from Note [Specialising Calls]: -- -- f :: forall a b c. Int -> Eq a => Show b => c -> Blah -- f @a @b @c i dEqA dShowA x = blah -- -- Suppose we decide to specialise it at the following pattern: -- -- [ SpecType T1, SpecType T2, UnspecType, UnspecArg -- , SpecDict dEqT1, SpecDict ($dfShow dShowT2), UnspecArg ] -- -- We'd eventually like to build the RULE -- -- RULE "SPEC f @T1 @T2 _" -- forall (@c :: Type) (i :: Int) (d1 :: Eq T1) (d2 :: Show T2). -- f @T1 @T2 @c i d1 d2 = $sf @c i -- -- and the specialisation '$sf' -- -- $sf :: forall c. Int -> c -> Blah -- $sf = SUBST[a :-> T1, b :-> T2, dEqA :-> dEqT1, dShowA :-> dShow1] (\@c i x -> blah) -- -- where dShow1 is a floated binding created by bindAuxiliaryDict. -- -- The cases for 'specHeader' below are presented in the same order as this -- running example. The result of 'specHeader' for this example is as follows: -- -- ( -- Returned arguments -- env + [a :-> T1, b :-> T2, dEqA :-> dEqT1, dShowA :-> dShow1] -- , [x] -- -- -- RULE helpers -- , [c, i, d1, d2] -- , [T1, T2, c, i, d1, d2] -- -- -- Specialised function helpers -- , [c, i, x] -- , [dShow1 = $dfShow dShowT2] -- , [T1, T2, c, i, dEqT1, dShow1] -- ) specHeader :: SpecEnv -> [InBndr] -- The binders from the original function 'f' -> [SpecArg] -- From the CallInfo -> SpecM ( Bool -- True <=> some useful specialisation happened -- Not the same as any (isSpecDict args) because -- the args might be longer than bndrs -- Returned arguments , SpecEnv -- Substitution to apply to the body of 'f' , [OutBndr] -- Leftover binders from the original function 'f' -- that don’t have a corresponding SpecArg -- RULE helpers , [OutBndr] -- Binders for the RULE , [OutExpr] -- Args for the LHS of the rule -- Specialised function helpers , [OutBndr] -- Binders for $sf , [DictBind] -- Auxiliary dictionary bindings , [OutExpr] -- Specialised arguments for unfolding -- Same length as "args for LHS of rule" ) -- We want to specialise on type 'T1', and so we must construct a substitution -- 'a->T1', as well as a LHS argument for the resulting RULE and unfolding -- details. specHeader env (bndr : bndrs) (SpecType t : args) = do { let env' = extendTvSubstList env [(bndr, t)] ; (useful, env'', leftover_bndrs, rule_bs, rule_es, bs', dx, spec_args) <- specHeader env' bndrs args ; pure ( useful , env'' , leftover_bndrs , rule_bs , Type t : rule_es , bs' , dx , Type t : spec_args ) } -- Next we have a type that we don't want to specialise. We need to perform -- a substitution on it (in case the type refers to 'a'). Additionally, we need -- to produce a binder, LHS argument and RHS argument for the resulting rule, -- /and/ a binder for the specialised body. specHeader env (bndr : bndrs) (UnspecType : args) = do { let (env', bndr') = substBndr env bndr ; (useful, env'', leftover_bndrs, rule_bs, rule_es, bs', dx, spec_args) <- specHeader env' bndrs args ; pure ( useful , env'' , leftover_bndrs , bndr' : rule_bs , varToCoreExpr bndr' : rule_es , bndr' : bs' , dx , varToCoreExpr bndr' : spec_args ) } -- Next we want to specialise the 'Eq a' dict away. We need to construct -- a wildcard binder to match the dictionary (See Note [Specialising Calls] for -- the nitty-gritty), as a LHS rule and unfolding details. specHeader env (bndr : bndrs) (SpecDict d : args) = do { bndr' <- newDictBndr env bndr -- See Note [Zap occ info in rule binders] ; let (env', dx_bind, spec_dict) = bindAuxiliaryDict env bndr bndr' d ; (_, env'', leftover_bndrs, rule_bs, rule_es, bs', dx, spec_args) <- specHeader env' bndrs args ; pure ( True -- Ha! A useful specialisation! , env'' , leftover_bndrs -- See Note [Evidence foralls] , exprFreeIdsList (varToCoreExpr bndr') ++ rule_bs , varToCoreExpr bndr' : rule_es , bs' , maybeToList dx_bind ++ dx , spec_dict : spec_args ) } -- Finally, we have the unspecialised argument 'i'. We need to produce -- a binder, LHS and RHS argument for the RULE, and a binder for the -- specialised body. -- -- NB: Calls to 'specHeader' will trim off any trailing 'UnspecArg's, which is -- why 'i' doesn't appear in our RULE above. But we have no guarantee that -- there aren't 'UnspecArg's which come /before/ all of the dictionaries, so -- this case must be here. specHeader env (bndr : bndrs) (UnspecArg : args) = do { -- see Note [Zap occ info in rule binders] let (env', bndr') = substBndr env (zapIdOccInfo bndr) ; (useful, env'', leftover_bndrs, rule_bs, rule_es, bs', dx, spec_args) <- specHeader env' bndrs args ; pure ( useful , env'' , leftover_bndrs , bndr' : rule_bs , varToCoreExpr bndr' : rule_es , if isDeadBinder bndr then bs' -- see Note [Drop dead args from specialisations] else bndr' : bs' , dx , varToCoreExpr bndr' : spec_args ) } -- If we run out of binders, stop immediately -- See Note [Specialisation Must Preserve Sharing] specHeader env [] _ = pure (False, env, [], [], [], [], [], []) -- Return all remaining binders from the original function. These have the -- invariant that they should all correspond to unspecialised arguments, so -- it's safe to stop processing at this point. specHeader env bndrs [] = pure (False, env', bndrs', [], [], [], [], []) where (env', bndrs') = substBndrs env bndrs -- | Binds a dictionary argument to a fresh name, to preserve sharing bindAuxiliaryDict :: SpecEnv -> InId -> OutId -> OutExpr -- Original dict binder, and the witnessing expression -> ( SpecEnv -- Substitute for orig_dict_id , Maybe DictBind -- Auxiliary dict binding, if any , OutExpr) -- Witnessing expression (always trivial) bindAuxiliaryDict env@(SE { se_subst = subst, se_interesting = interesting }) orig_dict_id fresh_dict_id dict_expr -- If the dictionary argument is trivial, -- don’t bother creating a new dict binding; just substitute | Just dict_id <- getIdFromTrivialExpr_maybe dict_expr = let env' = env { se_subst = Core.extendSubst subst orig_dict_id dict_expr `Core.extendInScope` dict_id -- See Note [Keep the old dictionaries interesting] , se_interesting = interesting `extendVarSet` dict_id } in (env', Nothing, dict_expr) | otherwise -- Non-trivial dictionary arg; make an auxiliary binding = let dict_bind = mkDB (NonRec fresh_dict_id dict_expr) env' = env { se_subst = Core.extendSubst subst orig_dict_id (Var fresh_dict_id) `Core.extendInScope` fresh_dict_id -- See Note [Make the new dictionaries interesting] , se_interesting = interesting `extendVarSet` fresh_dict_id } in (env', Just dict_bind, Var fresh_dict_id) {- Note [Make the new dictionaries interesting] ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Important! We're going to substitute dx_id1 for d and we want it to look "interesting", else we won't gather *any* consequential calls. E.g. f d = ...g d.... If we specialise f for a call (f (dfun dNumInt)), we'll get a consequent call (g d') with an auxiliary definition d' = df dNumInt We want that consequent call to look interesting Note [Keep the old dictionaries interesting] ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ In bindAuxiliaryDict, we don’t bother creating a new dict binding if the dict expression is trivial. For example, if we have f = \ @m1 (d1 :: Monad m1) -> ... and we specialize it at the pattern [SpecType IO, SpecArg $dMonadIO] it would be silly to create a new binding for $dMonadIO; it’s already a binding! So we just extend the substitution directly: m1 :-> IO d1 :-> $dMonadIO But this creates a new subtlety: the dict expression might be a dict binding we floated out while specializing another function. For example, we might have d2 = $p1Monad $dMonadIO -- floated out by bindAuxiliaryDict $sg = h @IO d2 h = \ @m2 (d2 :: Applicative m2) -> ... and end up specializing h at the following pattern: [SpecType IO, SpecArg d2] When we created the d2 binding in the first place, we locally marked it as interesting while specializing g as described above by Note [Make the new dictionaries interesting]. But when we go to specialize h, it isn’t in the SpecEnv anymore, so we’ve lost the knowledge that we should specialize on it. To fix this, we have to explicitly add d2 *back* to the interesting set. That way, it will still be considered interesting while specializing the body of h. See !2913. -} {- ********************************************************************* * * UsageDetails and suchlike * * ********************************************************************* -} data UsageDetails = MkUD { ud_binds :: !(Bag DictBind), -- See Note [Floated dictionary bindings] -- The order is important; -- in ds1 `union` ds2, bindings in ds2 can depend on those in ds1 -- (Remember, Bags preserve order in GHC.) ud_calls :: !CallDetails -- INVARIANT: suppose bs = bindersOf ud_binds -- Then 'calls' may *mention* 'bs', -- but there should be no calls *for* bs } -- | A 'DictBind' is a binding along with a cached set containing its free -- variables (both type variables and dictionaries) data DictBind = DB { db_bind :: CoreBind, db_fvs :: VarSet } {- Note [Floated dictionary bindings] ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ We float out dictionary bindings for the reasons described under "Dictionary floating" above. But not /just/ dictionary bindings. Consider f :: Eq a => blah f a d = rhs $c== :: T -> T -> Bool $c== x y = ... $df :: Eq T $df = Eq $c== ... gurgle = ...(f @T $df)... We gather the call info for (f @T $df), and we don't want to drop it when we come across the binding for $df. So we add $df to the floats and continue. But then we have to add $c== to the floats, and so on. These all float above the binding for 'f', and now we can successfully specialise 'f'. So the DictBinds in (ud_binds :: Bag DictBind) may contain non-dictionary bindings too. -} instance Outputable DictBind where ppr (DB { db_bind = bind, db_fvs = fvs }) = text "DB" <+> braces (sep [ text "bind:" <+> ppr bind , text "fvs: " <+> ppr fvs ]) instance Outputable UsageDetails where ppr (MkUD { ud_binds = dbs, ud_calls = calls }) = text "MkUD" <+> braces (sep (punctuate comma [text "binds" <+> equals <+> ppr dbs, text "calls" <+> equals <+> ppr calls])) emptyUDs :: UsageDetails emptyUDs = MkUD { ud_binds = emptyBag, ud_calls = emptyDVarEnv } ------------------------------------------------------------ type CallDetails = DIdEnv CallInfoSet -- The order of specialized binds and rules depends on how we linearize -- CallDetails, so to get determinism we must use a deterministic set here. -- See Note [Deterministic UniqFM] in GHC.Types.Unique.DFM data CallInfoSet = CIS Id (Bag CallInfo) -- The list of types and dictionaries is guaranteed to -- match the type of f -- The Bag may contain duplicate calls (i.e. f @T and another f @T) -- These dups are eliminated by already_covered in specCalls data CallInfo = CI { ci_key :: [SpecArg] -- All arguments , ci_fvs :: VarSet -- Free vars of the ci_key -- call (including tyvars) -- [*not* include the main id itself, of course] } type DictExpr = CoreExpr ciSetFilter :: (CallInfo -> Bool) -> CallInfoSet -> CallInfoSet ciSetFilter p (CIS id a) = CIS id (filterBag p a) instance Outputable CallInfoSet where ppr (CIS fn map) = hang (text "CIS" <+> ppr fn) 2 (ppr map) pprCallInfo :: Id -> CallInfo -> SDoc pprCallInfo fn (CI { ci_key = key }) = ppr fn <+> ppr key ppr_call_key_ty :: SpecArg -> Maybe SDoc ppr_call_key_ty (SpecType ty) = Just $ char '@' <> pprParendType ty ppr_call_key_ty UnspecType = Just $ char '_' ppr_call_key_ty (SpecDict _) = Nothing ppr_call_key_ty UnspecArg = Nothing instance Outputable CallInfo where ppr (CI { ci_key = key, ci_fvs = _fvs }) = text "CI" <> braces (sep (map ppr key)) unionCalls :: CallDetails -> CallDetails -> CallDetails unionCalls c1 c2 = plusDVarEnv_C unionCallInfoSet c1 c2 unionCallInfoSet :: CallInfoSet -> CallInfoSet -> CallInfoSet unionCallInfoSet (CIS f calls1) (CIS _ calls2) = CIS f (calls1 `unionBags` calls2) callDetailsFVs :: CallDetails -> VarSet callDetailsFVs calls = nonDetStrictFoldUDFM (unionVarSet . callInfoFVs) emptyVarSet calls -- It's OK to use nonDetStrictFoldUDFM here because we forget the ordering -- immediately by converting to a nondeterministic set. callInfoFVs :: CallInfoSet -> VarSet callInfoFVs (CIS _ call_info) = foldr (\(CI { ci_fvs = fv }) vs -> unionVarSet fv vs) emptyVarSet call_info getTheta :: [TyCoBinder] -> [PredType] getTheta = fmap tyBinderType . filter isInvisibleBinder . filter (not . isNamedBinder) ------------------------------------------------------------ singleCall :: Id -> [SpecArg] -> UsageDetails singleCall id args = MkUD {ud_binds = emptyBag, ud_calls = unitDVarEnv id $ CIS id $ unitBag (CI { ci_key = args -- used to be tys , ci_fvs = call_fvs }) } where call_fvs = foldr (unionVarSet . specArgFreeVars) emptyVarSet args -- The type args (tys) are guaranteed to be part of the dictionary -- types, because they are just the constrained types, -- and the dictionary is therefore sure to be bound -- inside the binding for any type variables free in the type; -- hence it's safe to neglect tyvars free in tys when making -- the free-var set for this call -- BUT I don't trust this reasoning; play safe and include tys_fvs -- -- We don't include the 'id' itself. mkCallUDs, mkCallUDs' :: SpecEnv -> Id -> [CoreExpr] -> UsageDetails mkCallUDs env f args = -- pprTrace "mkCallUDs" (vcat [ ppr f, ppr args, ppr res ]) res where res = mkCallUDs' env f args mkCallUDs' env f args | wantCallsFor env f -- We want it, and... , not (null ci_key) -- this call site has a useful specialisation = -- pprTrace "mkCallUDs: keeping" _trace_doc singleCall f ci_key | otherwise -- See also Note [Specialisations already covered] = -- pprTrace "mkCallUDs: discarding" _trace_doc emptyUDs where _trace_doc = vcat [ppr f, ppr args, ppr ci_key] pis = fst $ splitPiTys $ idType f constrained_tyvars = tyCoVarsOfTypes $ getTheta pis ci_key :: [SpecArg] ci_key = dropWhileEndLE (not . isSpecDict) $ zipWith mk_spec_arg args pis -- Drop trailing args until we get to a SpecDict -- In this way the RULE has as few args as possible, -- which broadens its applicability, since rules only -- fire when saturated mk_spec_arg :: CoreExpr -> TyCoBinder -> SpecArg mk_spec_arg arg (Named bndr) | binderVar bndr `elemVarSet` constrained_tyvars = case arg of Type ty -> SpecType ty _ -> pprPanic "ci_key" $ ppr arg | otherwise = UnspecType -- For "InvisArg", which are the type-class dictionaries, -- we decide on a case by case basis if we want to specialise -- on this argument; if so, SpecDict, if not UnspecArg mk_spec_arg arg (Anon InvisArg pred) | not (isIPLikePred (scaledThing pred)) -- See Note [Type determines value] , interestingDict env arg -- See Note [Interesting dictionary arguments] = SpecDict arg | otherwise = UnspecArg mk_spec_arg _ (Anon VisArg _) = UnspecArg wantCallsFor :: SpecEnv -> Id -> Bool wantCallsFor _env _f = True -- We could reduce the size of the UsageDetails by being less eager -- about collecting calls for LocalIds: there is no point for -- ones that are lambda-bound. We can't decide this by looking at -- the (absence of an) unfolding, because unfoldings for local -- functions are discarded by cloneBindSM, so no local binder will -- have an unfolding at this stage. We'd have to keep a candidate -- set of let-binders. -- -- Not many lambda-bound variables have dictionary arguments, so -- this would make little difference anyway. -- -- For imported Ids we could check for an unfolding, but we have to -- do so anyway in canSpecImport, and it seems better to have it -- all in one place. So we simply collect usage info for imported -- overloaded functions. {- Note [Type determines value] ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Only specialise on non-impicit-parameter predicates, because these are the ones whose *type* determines their *value*. In particular, with implicit params, the type args *don't* say what the value of the implicit param is! See #7101. So we treat implicit params just like ordinary arguments for the purposes of specialisation. Note that we still want to specialise functions with implicit params if they have *other* dicts which are class params; see #17930. Note [Interesting dictionary arguments] ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Consider this \a.\d:Eq a. let f = ... in ...(f d)... There really is not much point in specialising f wrt the dictionary d, because the code for the specialised f is not improved at all, because d is lambda-bound. We simply get junk specialisations. What is "interesting"? Just that it has *some* structure. But what about variables? * A variable might be imported, in which case its unfolding will tell us whether it has useful structure * Local variables are cloned on the way down (to avoid clashes when we float dictionaries), and cloning drops the unfolding (cloneIdBndr). Moreover, we make up some new bindings, and it's a nuisance to give them unfoldings. So we keep track of the "interesting" dictionaries as a VarSet in SpecEnv. We have to take care to put any new interesting dictionary bindings in the set. We accidentally lost accurate tracking of local variables for a long time, because cloned variables don't have unfoldings. But makes a massive difference in a few cases, eg #5113. For nofib as a whole it's only a small win: 2.2% improvement in allocation for ansi, 1.2% for bspt, but mostly 0.0! Average 0.1% increase in binary size. -} interestingDict :: SpecEnv -> CoreExpr -> Bool -- A dictionary argument is interesting if it has *some* structure -- NB: "dictionary" arguments include constraints of all sorts, -- including equality constraints; hence the Coercion case interestingDict env (Var v) = hasSomeUnfolding (idUnfolding v) || isDataConWorkId v || v `elemVarSet` se_interesting env interestingDict _ (Type _) = False interestingDict _ (Coercion _) = False interestingDict env (App fn (Type _)) = interestingDict env fn interestingDict env (App fn (Coercion _)) = interestingDict env fn interestingDict env (Tick _ a) = interestingDict env a interestingDict env (Cast e _) = interestingDict env e interestingDict _ _ = True plusUDs :: UsageDetails -> UsageDetails -> UsageDetails plusUDs (MkUD {ud_binds = db1, ud_calls = calls1}) (MkUD {ud_binds = db2, ud_calls = calls2}) = MkUD { ud_binds = db1 `unionBags` db2 , ud_calls = calls1 `unionCalls` calls2 } ----------------------------- _dictBindBndrs :: Bag DictBind -> [Id] _dictBindBndrs dbs = foldr ((++) . bindersOf . db_bind) [] dbs -- | Construct a 'DictBind' from a 'CoreBind' mkDB :: CoreBind -> DictBind mkDB bind = DB { db_bind = bind, db_fvs = bind_fvs bind } -- | Identify the free variables of a 'CoreBind' bind_fvs :: CoreBind -> VarSet bind_fvs (NonRec bndr rhs) = pair_fvs (bndr,rhs) bind_fvs (Rec prs) = foldl' delVarSet rhs_fvs bndrs where bndrs = map fst prs rhs_fvs = unionVarSets (map pair_fvs prs) pair_fvs :: (Id, CoreExpr) -> VarSet pair_fvs (bndr, rhs) = exprSomeFreeVars interesting rhs `unionVarSet` idFreeVars bndr -- idFreeVars: don't forget variables mentioned in -- the rules of the bndr. C.f. OccAnal.addRuleUsage -- Also tyvars mentioned in its type; they may not appear -- in the RHS -- type T a = Int -- x :: T a = 3 where interesting :: InterestingVarFun interesting v = isLocalVar v || (isId v && isDFunId v) -- Very important: include DFunIds /even/ if it is imported -- Reason: See Note [Avoiding loops], the second example -- involving an imported dfun. We must know whether -- a dictionary binding depends on an imported dfun, -- in case we try to specialise that imported dfun -- #13429 illustrates -- | Flatten a set of "dumped" 'DictBind's, and some other binding -- pairs, into a single recursive binding. recWithDumpedDicts :: [(Id,CoreExpr)] -> Bag DictBind -> DictBind recWithDumpedDicts pairs dbs = DB { db_bind = Rec bindings, db_fvs = fvs } where (bindings, fvs) = foldr add ([], emptyVarSet) (dbs `snocBag` mkDB (Rec pairs)) add (DB { db_bind = bind, db_fvs = fvs }) (prs_acc, fvs_acc) = case bind of NonRec b r -> ((b,r) : prs_acc, fvs') Rec prs1 -> (prs1 ++ prs_acc, fvs') where fvs' = fvs_acc `unionVarSet` fvs snocDictBinds :: UsageDetails -> [DictBind] -> UsageDetails -- Add ud_binds to the tail end of the bindings in uds snocDictBinds uds dbs = uds { ud_binds = ud_binds uds `unionBags` listToBag dbs } consDictBind :: DictBind -> UsageDetails -> UsageDetails consDictBind bind uds = uds { ud_binds = bind `consBag` ud_binds uds } addDictBinds :: [DictBind] -> UsageDetails -> UsageDetails addDictBinds binds uds = uds { ud_binds = listToBag binds `unionBags` ud_binds uds } snocDictBind :: UsageDetails -> DictBind -> UsageDetails snocDictBind uds bind = uds { ud_binds = ud_binds uds `snocBag` bind } wrapDictBinds :: Bag DictBind -> [CoreBind] -> [CoreBind] wrapDictBinds dbs binds = foldr add binds dbs where add (DB { db_bind = bind }) binds = bind : binds wrapDictBindsE :: Bag DictBind -> CoreExpr -> CoreExpr wrapDictBindsE dbs expr = foldr add expr dbs where add (DB { db_bind = bind }) expr = Let bind expr ---------------------- dumpUDs :: [CoreBndr] -> UsageDetails -> (UsageDetails, Bag DictBind) -- Used at a lambda or case binder; just dump anything mentioning the binder dumpUDs bndrs uds@(MkUD { ud_binds = orig_dbs, ud_calls = orig_calls }) | null bndrs = (uds, emptyBag) -- Common in case alternatives | otherwise = -- pprTrace "dumpUDs" (ppr bndrs $$ ppr free_uds $$ ppr dump_dbs) $ (free_uds, dump_dbs) where free_uds = MkUD { ud_binds = free_dbs, ud_calls = free_calls } bndr_set = mkVarSet bndrs (free_dbs, dump_dbs, dump_set) = splitDictBinds orig_dbs bndr_set free_calls = deleteCallsMentioning dump_set $ -- Drop calls mentioning bndr_set on the floor deleteCallsFor bndrs orig_calls -- Discard calls for bndr_set; there should be -- no calls for any of the dicts in dump_dbs dumpBindUDs :: [CoreBndr] -> UsageDetails -> (UsageDetails, Bag DictBind, Bool) -- Used at a let(rec) binding. -- We return a boolean indicating whether the binding itself is mentioned, -- directly or indirectly, by any of the ud_calls; in that case we want to -- float the binding itself; -- See Note [Floated dictionary bindings] dumpBindUDs bndrs (MkUD { ud_binds = orig_dbs, ud_calls = orig_calls }) = -- pprTrace "dumpBindUDs" (ppr bndrs $$ ppr free_uds $$ ppr dump_dbs) $ (free_uds, dump_dbs, float_all) where free_uds = MkUD { ud_binds = free_dbs, ud_calls = free_calls } bndr_set = mkVarSet bndrs (free_dbs, dump_dbs, dump_set) = splitDictBinds orig_dbs bndr_set free_calls = deleteCallsFor bndrs orig_calls float_all = dump_set `intersectsVarSet` callDetailsFVs free_calls callsForMe :: Id -> UsageDetails -> (UsageDetails, [CallInfo]) callsForMe fn (MkUD { ud_binds = orig_dbs, ud_calls = orig_calls }) = -- pprTrace ("callsForMe") -- (vcat [ppr fn, -- text "Orig dbs =" <+> ppr (_dictBindBndrs orig_dbs), -- text "Orig calls =" <+> ppr orig_calls, -- text "Dep set =" <+> ppr dep_set, -- text "Calls for me =" <+> ppr calls_for_me]) $ (uds_without_me, calls_for_me) where uds_without_me = MkUD { ud_binds = orig_dbs , ud_calls = delDVarEnv orig_calls fn } calls_for_me = case lookupDVarEnv orig_calls fn of Nothing -> [] Just cis -> filterCalls cis orig_dbs -- filterCalls: drop calls that (directly or indirectly) -- refer to fn. See Note [Avoiding loops] ---------------------- filterCalls :: CallInfoSet -> Bag DictBind -> [CallInfo] -- See Note [Avoiding loops] filterCalls (CIS fn call_bag) dbs = filter ok_call (bagToList call_bag) where dump_set = foldl' go (unitVarSet fn) dbs -- This dump-set could also be computed by splitDictBinds -- (_,_,dump_set) = splitDictBinds dbs {fn} -- But this variant is shorter go so_far (DB { db_bind = bind, db_fvs = fvs }) | fvs `intersectsVarSet` so_far = extendVarSetList so_far (bindersOf bind) | otherwise = so_far ok_call (CI { ci_fvs = fvs }) = fvs `disjointVarSet` dump_set ---------------------- splitDictBinds :: Bag DictBind -> IdSet -> (Bag DictBind, Bag DictBind, IdSet) -- splitDictBinds dbs bndrs returns -- (free_dbs, dump_dbs, dump_set) -- where -- * dump_dbs depends, transitively on bndrs -- * free_dbs does not depend on bndrs -- * dump_set = bndrs `union` bndrs(dump_dbs) splitDictBinds dbs bndr_set = foldl' split_db (emptyBag, emptyBag, bndr_set) dbs -- Important that it's foldl' not foldr; -- we're accumulating the set of dumped ids in dump_set where split_db (free_dbs, dump_dbs, dump_idset) db | DB { db_bind = bind, db_fvs = fvs } <- db , dump_idset `intersectsVarSet` fvs -- Dump it = (free_dbs, dump_dbs `snocBag` db, extendVarSetList dump_idset (bindersOf bind)) | otherwise -- Don't dump it = (free_dbs `snocBag` db, dump_dbs, dump_idset) ---------------------- deleteCallsMentioning :: VarSet -> CallDetails -> CallDetails -- Remove calls *mentioning* bs in any way deleteCallsMentioning bs calls = mapDVarEnv (ciSetFilter keep_call) calls where keep_call (CI { ci_fvs = fvs }) = fvs `disjointVarSet` bs deleteCallsFor :: [Id] -> CallDetails -> CallDetails -- Remove calls *for* bs deleteCallsFor bs calls = delDVarEnvList calls bs {- ************************************************************************ * * \subsubsection{Boring helper functions} * * ************************************************************************ -} type SpecM a = UniqSM a runSpecM :: SpecM a -> CoreM a runSpecM thing_inside = do { us <- getUniqueSupplyM ; return (initUs_ us thing_inside) } mapAndCombineSM :: (a -> SpecM (b, UsageDetails)) -> [a] -> SpecM ([b], UsageDetails) mapAndCombineSM _ [] = return ([], emptyUDs) mapAndCombineSM f (x:xs) = do (y, uds1) <- f x (ys, uds2) <- mapAndCombineSM f xs return (y:ys, uds1 `plusUDs` uds2) extendTvSubstList :: SpecEnv -> [(TyVar,Type)] -> SpecEnv extendTvSubstList env tv_binds = env { se_subst = Core.extendTvSubstList (se_subst env) tv_binds } substTy :: SpecEnv -> Type -> Type substTy env ty = Core.substTy (se_subst env) ty substCo :: SpecEnv -> Coercion -> Coercion substCo env co = Core.substCo (se_subst env) co substBndr :: SpecEnv -> CoreBndr -> (SpecEnv, CoreBndr) substBndr env bs = case Core.substBndr (se_subst env) bs of (subst', bs') -> (env { se_subst = subst' }, bs') substBndrs :: SpecEnv -> [CoreBndr] -> (SpecEnv, [CoreBndr]) substBndrs env bs = case Core.substBndrs (se_subst env) bs of (subst', bs') -> (env { se_subst = subst' }, bs') cloneBindSM :: SpecEnv -> CoreBind -> SpecM (SpecEnv, SpecEnv, CoreBind) -- Clone the binders of the bind; return new bind with the cloned binders -- Return the substitution to use for RHSs, and the one to use for the body cloneBindSM env@(SE { se_subst = subst, se_interesting = interesting }) (NonRec bndr rhs) = do { us <- getUniqueSupplyM ; let (subst', bndr') = Core.cloneIdBndr subst us bndr interesting' | interestingDict env rhs = interesting `extendVarSet` bndr' | otherwise = interesting ; return (env, env { se_subst = subst', se_interesting = interesting' } , NonRec bndr' rhs) } cloneBindSM env@(SE { se_subst = subst, se_interesting = interesting }) (Rec pairs) = do { us <- getUniqueSupplyM ; let (subst', bndrs') = Core.cloneRecIdBndrs subst us (map fst pairs) env' = env { se_subst = subst' , se_interesting = interesting `extendVarSetList` [ v | (v,r) <- pairs, interestingDict env r ] } ; return (env', env', Rec (bndrs' `zip` map snd pairs)) } newDictBndr :: SpecEnv -> CoreBndr -> SpecM CoreBndr -- Make up completely fresh binders for the dictionaries -- Their bindings are going to float outwards newDictBndr env b = do { uniq <- getUniqueM ; let n = idName b ty' = substTy env (idType b) ; return (mkUserLocal (nameOccName n) uniq Many ty' (getSrcSpan n)) } newSpecIdSM :: Id -> Type -> Maybe JoinArity -> SpecM Id -- Give the new Id a similar occurrence name to the old one newSpecIdSM old_id new_ty join_arity_maybe = do { uniq <- getUniqueM ; let name = idName old_id new_occ = mkSpecOcc (nameOccName name) new_id = mkUserLocal new_occ uniq Many new_ty (getSrcSpan name) `asJoinId_maybe` join_arity_maybe ; return new_id } {- Old (but interesting) stuff about unboxed bindings ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ What should we do when a value is specialised to a *strict* unboxed value? map_*_* f (x:xs) = let h = f x t = map f xs in h:t Could convert let to case: map_*_Int# f (x:xs) = case f x of h# -> let t = map f xs in h#:t This may be undesirable since it forces evaluation here, but the value may not be used in all branches of the body. In the general case this transformation is impossible since the mutual recursion in a letrec cannot be expressed as a case. There is also a problem with top-level unboxed values, since our implementation cannot handle unboxed values at the top level. Solution: Lift the binding of the unboxed value and extract it when it is used: map_*_Int# f (x:xs) = let h = case (f x) of h# -> _Lift h# t = map f xs in case h of _Lift h# -> h#:t Now give it to the simplifier and the _Lifting will be optimised away. The benefit is that we have given the specialised "unboxed" values a very simple lifted semantics and then leave it up to the simplifier to optimise it --- knowing that the overheads will be removed in nearly all cases. In particular, the value will only be evaluated in the branches of the program which use it, rather than being forced at the point where the value is bound. For example: filtermap_*_* p f (x:xs) = let h = f x t = ... in case p x of True -> h:t False -> t ==> filtermap_*_Int# p f (x:xs) = let h = case (f x) of h# -> _Lift h# t = ... in case p x of True -> case h of _Lift h# -> h#:t False -> t The binding for h can still be inlined in the one branch and the _Lifting eliminated. Question: When won't the _Lifting be eliminated? Answer: When they at the top-level (where it is necessary) or when inlining would duplicate work (or possibly code depending on options). However, the _Lifting will still be eliminated if the strictness analyser deems the lifted binding strict. -}