{-
(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

import GHC.Prelude

import GHC.Driver.DynFlags
import GHC.Driver.Config
import GHC.Driver.Config.Diagnostic
import GHC.Driver.Config.Core.Rules ( initRuleOpts )

import GHC.Core.Type  hiding( substTy, substCo, extendTvSubst, zapSubst )
import GHC.Core.Multiplicity
import GHC.Core.SimpleOpt( defaultSimpleOpts, simpleOptExprWith )
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.Make      ( mkLitRubbish )
import GHC.Core.Unify     ( tcMatchTy )
import GHC.Core.Rules
import GHC.Core.Utils     ( exprIsTrivial, exprIsTopLevelBindable
                          , mkCast, exprType
                          , stripTicksTop, mkInScopeSetBndrs )
import GHC.Core.FVs
import GHC.Core.TyCo.FVs ( tyCoVarsOfTypeList )
import GHC.Core.Opt.Arity( collectBindersPushingCo )
-- import GHC.Core.Ppr( pprIds )

import GHC.Builtin.Types  ( unboxedUnitTy )

import GHC.Data.Maybe     ( maybeToList, isJust )
import GHC.Data.Bag
import GHC.Data.OrdList
import GHC.Data.List.SetOps

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      ( PiTyBinder(..), isLocalVar, isInvisibleFunArg, mkLocalVar )
import GHC.Types.Var.Set
import GHC.Types.Var.Env
import GHC.Types.Id
import GHC.Types.Id.Info
import GHC.Types.Error

import GHC.Utils.Error ( mkMCDiagnostic )
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.Core.Unfold

import Data.List( partition )
import Data.List.NonEmpty ( NonEmpty (..) )
import GHC.Core.Subst (substTickish)

{-
************************************************************************
*                                                                      *
\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 specialisation 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)] => ...


Note [Specialisation and overlapping instances]
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Here is at tricky case (see a comment in MR !8916):

    module A where
      class C a where
        meth :: a -> String
      instance {-# OVERLAPPABLE #-} C (Maybe a) where
        meth _ = "Maybe"

      {-# SPECIALISE f :: Maybe a -> Bool -> String #-}
      f :: C a => a -> Bool -> String
      f a True = f a False
      f a _    = meth a

    module B where
      import A

      instance C (Maybe Int) where
        meth _ = "Int"

      main = putStrLn $ f (Just 42 :: Maybe Int) True

Running main without optimisations yields "Int", the correct answer.
Activating optimisations yields "Maybe" due to a rewrite rule in module
A generated by the SPECIALISE pragma:

    RULE "USPEC f" forall a (d :: C a). f @a d = $sf

In B we get the call (f @(Maybe Int) (d :: C (Maybe Int))), and
that rewrites to $sf, but that isn't really right.

Overlapping instances mean that `C (Maybe Int)` is not a singleton
type: there two distinct dictionaries that have this type.  And that
spells trouble for specialistion, which really asssumes singleton
types.

For now, we just accept this problem, but it may bite us one day.
One solution would be to decline to expose any specialisation rules
to an importing module -- but that seems a bit drastic.


************************************************************************
*                                                                      *
\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 :: ModGuts -> CoreM ModGuts
specProgram guts :: ModGuts
guts@(ModGuts { mg_module :: ModGuts -> Module
mg_module = Module
this_mod
                          , mg_rules :: ModGuts -> [CoreRule]
mg_rules  = [CoreRule]
local_rules
                          , mg_binds :: ModGuts -> CoreProgram
mg_binds  = CoreProgram
binds })
  = do { dflags   <- CoreM DynFlags
forall (m :: * -> *). HasDynFlags m => m DynFlags
getDynFlags
       ; rule_env <- initRuleEnv guts
                     -- See Note [Fire rules in the specialiser]

              -- 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 :: Subst
se_subst = InScopeSet -> Subst
Core.mkEmptySubst (InScopeSet -> Subst) -> InScopeSet -> Subst
forall a b. (a -> b) -> a -> b
$
                                        CoreProgram -> InScopeSet
mkInScopeSetBndrs CoreProgram
binds
                                      --    mkInScopeSetList $
                                      --  bindersOfBinds binds
                          , se_module :: Module
se_module = Module
this_mod
                          , se_rules :: RuleEnv
se_rules  = RuleEnv
rule_env
                          , se_dflags :: DynFlags
se_dflags = DynFlags
dflags }

             go []           = (CoreProgram, UsageDetails) -> SpecM (CoreProgram, UsageDetails)
forall a. a -> UniqSM a
forall (m :: * -> *) a. Monad m => a -> m a
return ([], UsageDetails
emptyUDs)
             go (InBind
bind:CoreProgram
binds) = do (bind', binds', uds') <- TopLevelFlag
-> SpecEnv
-> InBind
-> (SpecEnv -> SpecM (CoreProgram, UsageDetails))
-> SpecM (CoreProgram, CoreProgram, UsageDetails)
forall body.
TopLevelFlag
-> SpecEnv
-> InBind
-> (SpecEnv -> SpecM (body, UsageDetails))
-> SpecM (CoreProgram, body, UsageDetails)
specBind TopLevelFlag
TopLevel SpecEnv
top_env InBind
bind ((SpecEnv -> SpecM (CoreProgram, UsageDetails))
 -> SpecM (CoreProgram, CoreProgram, UsageDetails))
-> (SpecEnv -> SpecM (CoreProgram, UsageDetails))
-> SpecM (CoreProgram, CoreProgram, UsageDetails)
forall a b. (a -> b) -> a -> b
$ \SpecEnv
_ ->
                                                           CoreProgram -> SpecM (CoreProgram, UsageDetails)
go CoreProgram
binds
                                  return (bind' ++ binds', uds')

             -- Specialise the bindings of this module
       ; (binds', uds) <- runSpecM (go binds)

       ; (spec_rules, spec_binds) <- specImports top_env 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
*                                                                      *
********************************************************************* -}

{- Note [Specialising imported functions]
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
specImports specialises imported functions, based on calls in this module.

When -fspecialise-aggressively is on, we specialise any imported
function for which we have an unfolding.  The
-fspecialise-aggressively flag is usually off, because we risk lots of
orphan modules from over-vigorous specialisation.  (See Note [Orphans]
in GHC.Core.) However it's not a big deal: anything non-recursive with
an unfolding-template will probably have been inlined already.

When -fspecialise-aggressively is off, we are more selective about
specialisation (see canSpecImport):

(1) Without -fspecialise-aggressively, do not specialise
    DFunUnfoldings. Note [Do not specialise imported DFuns].

(2) Without -fspecialise-aggressively, specialise only imported things
    that have a /user-supplied/ INLINE or INLINABLE pragma (hence
    isAnyInlinePragma rather than isStableSource).

    In particular, we don't want to specialise workers created by
    worker/wrapper (for functions with no pragma) because they won't
    specialise usefully, and they generate quite a bit of useless code
    bloat.

    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

Wrinkle (W1): If we specialise an imported Id M.foo, we make a /local/
binding $sfoo.  But specImports may further specialise $sfoo. So we end up
with RULES for both M.foo (imported) and $sfoo (local).  Rules for local
Ids should be attached to the Ids themselves (see GHC.HsToCore
Note [Attach rules to local ids]); so we must partition the rules and
attach the local rules.  That is done in specImports, via addRulesToId.

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.
-}

specImports :: SpecEnv
            -> UsageDetails
            -> CoreM ([CoreRule], [CoreBind])
specImports :: SpecEnv -> UsageDetails -> CoreM ([CoreRule], CoreProgram)
specImports SpecEnv
top_env (MkUD { ud_binds :: UsageDetails -> FloatedDictBinds
ud_binds = FloatedDictBinds
dict_binds, ud_calls :: UsageDetails -> CallDetails
ud_calls = CallDetails
calls })
  | Bool -> Bool
not (Bool -> Bool) -> Bool -> Bool
forall a b. (a -> b) -> a -> b
$ GeneralFlag -> DynFlags -> Bool
gopt GeneralFlag
Opt_CrossModuleSpecialise (SpecEnv -> DynFlags
se_dflags SpecEnv
top_env)
    -- See Note [Disabling cross-module specialisation]
  = ([CoreRule], CoreProgram) -> CoreM ([CoreRule], CoreProgram)
forall a. a -> CoreM a
forall (m :: * -> *) a. Monad m => a -> m a
return ([], FloatedDictBinds -> CoreProgram -> CoreProgram
wrapDictBinds FloatedDictBinds
dict_binds [])

  | Bool
otherwise
  = do { let env_w_dict_bndrs :: SpecEnv
env_w_dict_bndrs = SpecEnv
top_env SpecEnv -> FloatedDictBinds -> SpecEnv
`bringFloatedDictsIntoScope` FloatedDictBinds
dict_binds
       ; (_env, spec_rules, spec_binds) <- SpecEnv
-> [Id]
-> FloatedDictBinds
-> CallDetails
-> CoreM (SpecEnv, [CoreRule], CoreProgram)
spec_imports SpecEnv
env_w_dict_bndrs [] FloatedDictBinds
dict_binds CallDetails
calls

             -- Make a Rec: see Note [Glom the bindings if imported functions are specialised]
             --
             -- wrapDictBinds: don't forget to wrap the specialized bindings with
             --   bindings for the needed dictionaries.
             --   See Note [Wrap bindings returned by specImports]
             --
             -- addRulesToId: see Wrinkle (W1) in Note [Specialising imported functions]
             --               c.f. GHC.HsToCore.addExportFlagsAndRules
       ; let (rules_for_locals, rules_for_imps) = partition isLocalRule spec_rules
             local_rule_base = RuleBase -> [CoreRule] -> RuleBase
extendRuleBaseList RuleBase
emptyRuleBase [CoreRule]
rules_for_locals
             final_binds
               | CoreProgram -> Bool
forall a. [a] -> Bool
forall (t :: * -> *) a. Foldable t => t a -> Bool
null CoreProgram
spec_binds = FloatedDictBinds -> CoreProgram -> CoreProgram
wrapDictBinds FloatedDictBinds
dict_binds []
               | Bool
otherwise       = [[(Id, OutExpr)] -> InBind
forall b. [(b, Expr b)] -> Bind b
Rec ([(Id, OutExpr)] -> InBind) -> [(Id, OutExpr)] -> InBind
forall a b. (a -> b) -> a -> b
$ (Id -> Id) -> [(Id, OutExpr)] -> [(Id, OutExpr)]
forall (f :: * -> *) a c b.
Functor f =>
(a -> c) -> f (a, b) -> f (c, b)
mapFst (RuleBase -> Id -> Id
addRulesToId RuleBase
local_rule_base) ([(Id, OutExpr)] -> [(Id, OutExpr)])
-> [(Id, OutExpr)] -> [(Id, OutExpr)]
forall a b. (a -> b) -> a -> b
$
                                          CoreProgram -> [(Id, OutExpr)]
forall b. [Bind b] -> [(b, Expr b)]
flattenBinds                          (CoreProgram -> [(Id, OutExpr)]) -> CoreProgram -> [(Id, OutExpr)]
forall a b. (a -> b) -> a -> b
$
                                          FloatedDictBinds -> CoreProgram -> CoreProgram
wrapDictBinds FloatedDictBinds
dict_binds              (CoreProgram -> CoreProgram) -> CoreProgram -> CoreProgram
forall a b. (a -> b) -> a -> b
$
                                          CoreProgram
spec_binds]

       ; return (rules_for_imps, final_binds)
    }

-- | Specialise a set of calls to imported bindings
spec_imports :: SpecEnv          -- Passed in so that all top-level Ids are in scope
                                 ---In-scope set includes the FloatedDictBinds
             -> [Id]             -- Stack of imported functions being specialised
                                 -- See Note [specImport call stack]
             -> FloatedDictBinds -- Dict bindings, used /only/ for filterCalls
                                 -- See Note [Avoiding loops in specImports]
             -> CallDetails      -- Calls for imported things
             -> CoreM ( SpecEnv      -- Env contains the new rules
                      , [CoreRule]   -- New rules
                      , [CoreBind] ) -- Specialised bindings
spec_imports :: SpecEnv
-> [Id]
-> FloatedDictBinds
-> CallDetails
-> CoreM (SpecEnv, [CoreRule], CoreProgram)
spec_imports SpecEnv
env [Id]
callers FloatedDictBinds
dict_binds CallDetails
calls
  = do { let import_calls :: [CallInfoSet]
import_calls = CallDetails -> [CallInfoSet]
forall a. DVarEnv a -> [a]
dVarEnvElts CallDetails
calls
--       ; debugTraceMsg (text "specImports {" <+>
--                         vcat [ text "calls:" <+> ppr import_calls
--                              , text "dict_binds:" <+> ppr dict_binds ])
       ; (env, rules, spec_binds) <- SpecEnv
-> [CallInfoSet] -> CoreM (SpecEnv, [CoreRule], CoreProgram)
go SpecEnv
env [CallInfoSet]
import_calls
--       ; debugTraceMsg (text "End specImports }" <+> ppr import_calls)

       ; return (env, rules, spec_binds) }
  where
    go :: SpecEnv -> [CallInfoSet] -> CoreM (SpecEnv, [CoreRule], [CoreBind])
    go :: SpecEnv
-> [CallInfoSet] -> CoreM (SpecEnv, [CoreRule], CoreProgram)
go SpecEnv
env [] = (SpecEnv, [CoreRule], CoreProgram)
-> CoreM (SpecEnv, [CoreRule], CoreProgram)
forall a. a -> CoreM a
forall (m :: * -> *) a. Monad m => a -> m a
return (SpecEnv
env, [], [])
    go SpecEnv
env (CallInfoSet
cis : [CallInfoSet]
other_calls)
      = do { -- debugTraceMsg (text "specImport {" <+> ppr cis)
           ; (env, rules1, spec_binds1) <- SpecEnv
-> [Id]
-> FloatedDictBinds
-> CallInfoSet
-> CoreM (SpecEnv, [CoreRule], CoreProgram)
spec_import SpecEnv
env [Id]
callers FloatedDictBinds
dict_binds CallInfoSet
cis
           ; -- debugTraceMsg (text "specImport }" <+> ppr cis)

           ; (env, rules2, spec_binds2) <- go env other_calls
           ; return (env, rules1 ++ rules2, spec_binds1 ++ spec_binds2) }

spec_import :: SpecEnv               -- Passed in so that all top-level Ids are in scope
                                     ---In-scope set includes the FloatedDictBinds
            -> [Id]                  -- Stack of imported functions being specialised
                                     -- See Note [specImport call stack]
            -> FloatedDictBinds      -- Dict bindings, used /only/ for filterCalls
                                     -- See Note [Avoiding loops in specImports]
            -> CallInfoSet           -- Imported function and calls for it
            -> CoreM ( SpecEnv
                     , [CoreRule]    -- New rules
                     , [CoreBind] )  -- Specialised bindings
spec_import :: SpecEnv
-> [Id]
-> FloatedDictBinds
-> CallInfoSet
-> CoreM (SpecEnv, [CoreRule], CoreProgram)
spec_import SpecEnv
env [Id]
callers FloatedDictBinds
dict_binds cis :: CallInfoSet
cis@(CIS Id
fn Bag CallInfo
_)
  | String -> Id -> [Id] -> Bool
forall a. Eq a => String -> a -> [a] -> Bool
isIn String
"specImport" Id
fn [Id]
callers
  = (SpecEnv, [CoreRule], CoreProgram)
-> CoreM (SpecEnv, [CoreRule], CoreProgram)
forall a. a -> CoreM a
forall (m :: * -> *) a. Monad m => a -> m a
return (SpecEnv
env, [], [])  -- 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

  | [CallInfo] -> Bool
forall a. [a] -> Bool
forall (t :: * -> *) a. Foldable t => t a -> Bool
null [CallInfo]
good_calls
  = (SpecEnv, [CoreRule], CoreProgram)
-> CoreM (SpecEnv, [CoreRule], CoreProgram)
forall a. a -> CoreM a
forall (m :: * -> *) a. Monad m => a -> m a
return (SpecEnv
env, [], [])

  | Just OutExpr
rhs <- DynFlags -> Id -> Maybe OutExpr
canSpecImport DynFlags
dflags Id
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
       ; eps_rules <- CoreM RuleBase
getExternalRuleBase
       ; let rule_env = SpecEnv -> RuleEnv
se_rules SpecEnv
env RuleEnv -> RuleBase -> RuleEnv
`updExternalPackageRules` RuleBase
eps_rules

--       ; debugTraceMsg (text "specImport1" <+> vcat
--           [ text "function:" <+> ppr fn
--           , text "good calls:" <+> ppr good_calls
--           , text "existing rules:" <+> ppr (getRules rule_env fn)
--           , text "rhs:" <+> ppr rhs
--           , text "dict_binds:" <+> ppr dict_binds ])

       ; (rules1, spec_pairs, MkUD { ud_binds = dict_binds1, ud_calls = new_calls })
            <- runSpecM $ specCalls True env (getRules rule_env fn) good_calls fn rhs

       ; let spec_binds1 = [Id -> OutExpr -> InBind
forall b. b -> Expr b -> Bind b
NonRec Id
b OutExpr
r | (Id
b,OutExpr
r) <- [(Id, OutExpr)]
spec_pairs]
             -- After the rules kick in, via fireRewriteRules, 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]
             -- Meanwhile, though, bring the binders into scope

             new_subst = SpecEnv -> Subst
se_subst SpecEnv
env Subst -> [Id] -> Subst
`Core.extendSubstInScopeList` ((Id, OutExpr) -> Id) -> [(Id, OutExpr)] -> [Id]
forall a b. (a -> b) -> [a] -> [b]
map (Id, OutExpr) -> Id
forall a b. (a, b) -> a
fst [(Id, OutExpr)]
spec_pairs
             new_env   = SpecEnv
env { se_rules = rule_env `addLocalRules` rules1
                             , se_subst = new_subst }
                         SpecEnv -> FloatedDictBinds -> SpecEnv
`bringFloatedDictsIntoScope` FloatedDictBinds
dict_binds1

       -- Now specialise any cascaded calls
--       ; debugTraceMsg (text "specImport 2" <+> vcat
--           [ text "function:" <+> ppr fn
--           , text "rules1:" <+> ppr rules1
--           , text "spec_binds1" <+> ppr spec_binds1
--           , text "dict_binds1" <+> ppr dict_binds1
--           , text "new_calls" <+> ppr new_calls ])

       ; (env, rules2, spec_binds2)
            <- spec_imports new_env (fn:callers)
                                    (dict_binds `thenFDBs` dict_binds1)
                                    new_calls

       ; let final_binds = FloatedDictBinds -> CoreProgram -> CoreProgram
wrapDictBinds FloatedDictBinds
dict_binds1 (CoreProgram -> CoreProgram) -> CoreProgram -> CoreProgram
forall a b. (a -> b) -> a -> b
$
                           CoreProgram
spec_binds2 CoreProgram -> CoreProgram -> CoreProgram
forall a. [a] -> [a] -> [a]
++ CoreProgram
spec_binds1

       ; return (env, rules2 ++ rules1, final_binds) }

  | Bool
otherwise
  = do { DynFlags -> [Id] -> Id -> [CallInfo] -> CoreM ()
tryWarnMissingSpecs DynFlags
dflags [Id]
callers Id
fn [CallInfo]
good_calls
       ; (SpecEnv, [CoreRule], CoreProgram)
-> CoreM (SpecEnv, [CoreRule], CoreProgram)
forall a. a -> CoreM a
forall (m :: * -> *) a. Monad m => a -> m a
return (SpecEnv
env, [], [])}

  where
    dflags :: DynFlags
dflags = SpecEnv -> DynFlags
se_dflags SpecEnv
env
    good_calls :: [CallInfo]
good_calls = CallInfoSet -> FloatedDictBinds -> [CallInfo]
filterCalls CallInfoSet
cis FloatedDictBinds
dict_binds
       -- SUPER IMPORTANT!  Drop calls that (directly or indirectly) refer to fn
       -- See Note [Avoiding loops in specImports]

canSpecImport :: DynFlags -> Id -> Maybe CoreExpr
canSpecImport :: DynFlags -> Id -> Maybe OutExpr
canSpecImport DynFlags
dflags Id
fn
  | Id -> Bool
isDataConWrapId Id
fn
  = Maybe OutExpr
forall a. Maybe a
Nothing   -- Don't specialise data-con wrappers, even if they
              -- have dict args; there is no benefit.

  | CoreUnfolding { uf_tmpl :: Unfolding -> OutExpr
uf_tmpl = OutExpr
rhs } <- Unfolding
unf
    -- CoreUnfolding: see Note [Specialising imported functions] point (1).
  , InlinePragma -> Bool
isAnyInlinePragma (Id -> InlinePragma
idInlinePragma Id
fn)
    -- See Note [Specialising imported functions] point (2).
  = OutExpr -> Maybe OutExpr
forall a. a -> Maybe a
Just OutExpr
rhs

  | GeneralFlag -> DynFlags -> Bool
gopt GeneralFlag
Opt_SpecialiseAggressively DynFlags
dflags
  = Unfolding -> Maybe OutExpr
maybeUnfoldingTemplate Unfolding
unf
    -- With -fspecialise-aggressively, specialise anything
    -- with an unfolding, stable or not, DFun or not

  | Bool
otherwise = Maybe OutExpr
forall a. Maybe a
Nothing
  where
    unf :: Unfolding
unf = Id -> Unfolding
realIdUnfolding Id
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 :: DynFlags -> [Id] -> Id -> [CallInfo] -> CoreM ()
tryWarnMissingSpecs DynFlags
dflags [Id]
callers Id
fn [CallInfo]
calls_for_fn
  | Id -> Bool
isClassOpId Id
fn = () -> CoreM ()
forall a. a -> CoreM a
forall (m :: * -> *) a. Monad m => a -> m a
return () -- See Note [Missed specialisation for ClassOps]
  | WarningFlag -> DynFlags -> Bool
wopt WarningFlag
Opt_WarnMissedSpecs DynFlags
dflags
    Bool -> Bool -> Bool
&& Bool -> Bool
not ([Id] -> Bool
forall a. [a] -> Bool
forall (t :: * -> *) a. Foldable t => t a -> Bool
null [Id]
callers)
    Bool -> Bool -> Bool
&& Bool
allCallersInlined                  = DiagnosticReason -> CoreM ()
doWarn (DiagnosticReason -> CoreM ()) -> DiagnosticReason -> CoreM ()
forall a b. (a -> b) -> a -> b
$ WarningFlag -> DiagnosticReason
WarningWithFlag WarningFlag
Opt_WarnMissedSpecs
  | WarningFlag -> DynFlags -> Bool
wopt WarningFlag
Opt_WarnAllMissedSpecs DynFlags
dflags    = DiagnosticReason -> CoreM ()
doWarn (DiagnosticReason -> CoreM ()) -> DiagnosticReason -> CoreM ()
forall a b. (a -> b) -> a -> b
$ WarningFlag -> DiagnosticReason
WarningWithFlag WarningFlag
Opt_WarnAllMissedSpecs
  | Bool
otherwise                             = () -> CoreM ()
forall a. a -> CoreM a
forall (m :: * -> *) a. Monad m => a -> m a
return ()
  where
    allCallersInlined :: Bool
allCallersInlined = (Id -> Bool) -> [Id] -> Bool
forall (t :: * -> *) a. Foldable t => (a -> Bool) -> t a -> Bool
all (InlinePragma -> Bool
isAnyInlinePragma (InlinePragma -> Bool) -> (Id -> InlinePragma) -> Id -> Bool
forall b c a. (b -> c) -> (a -> b) -> a -> c
. Id -> InlinePragma
idInlinePragma) [Id]
callers
    diag_opts :: DiagOpts
diag_opts = DynFlags -> DiagOpts
initDiagOpts DynFlags
dflags
    doWarn :: DiagnosticReason -> CoreM ()
doWarn DiagnosticReason
reason =
      MessageClass -> SDoc -> CoreM ()
msg (DiagOpts
-> DiagnosticReason -> Maybe DiagnosticCode -> MessageClass
mkMCDiagnostic DiagOpts
diag_opts DiagnosticReason
reason Maybe DiagnosticCode
forall a. Maybe a
Nothing)
        ([SDoc] -> SDoc
forall doc. IsDoc doc => [doc] -> doc
vcat [ SDoc -> Int -> SDoc -> SDoc
hang (String -> SDoc
forall doc. IsLine doc => String -> doc
text (String
"Could not specialise imported function") SDoc -> SDoc -> SDoc
forall doc. IsLine doc => doc -> doc -> doc
<+> SDoc -> SDoc
quotes (Id -> SDoc
forall a. Outputable a => a -> SDoc
ppr Id
fn))
                Int
2 ([SDoc] -> SDoc
forall doc. IsDoc doc => [doc] -> doc
vcat [ String -> SDoc
forall doc. IsLine doc => String -> doc
text String
"when specialising" SDoc -> SDoc -> SDoc
forall doc. IsLine doc => doc -> doc -> doc
<+> SDoc -> SDoc
quotes (Id -> SDoc
forall a. Outputable a => a -> SDoc
ppr Id
caller)
                        | Id
caller <- [Id]
callers])
          , SDoc -> SDoc
forall doc. IsOutput doc => doc -> doc
whenPprDebug (String -> SDoc
forall doc. IsLine doc => String -> doc
text String
"calls:" SDoc -> SDoc -> SDoc
forall doc. IsLine doc => doc -> doc -> doc
<+> [SDoc] -> SDoc
forall doc. IsDoc doc => [doc] -> doc
vcat ((CallInfo -> SDoc) -> [CallInfo] -> [SDoc]
forall a b. (a -> b) -> [a] -> [b]
map (Id -> CallInfo -> SDoc
pprCallInfo Id
fn) [CallInfo]
calls_for_fn))
          , String -> SDoc
forall doc. IsLine doc => String -> doc
text String
"Probable fix: add INLINABLE pragma on" SDoc -> SDoc -> SDoc
forall doc. IsLine doc => doc -> doc -> doc
<+> SDoc -> SDoc
quotes (Id -> SDoc
forall a. Outputable a => a -> SDoc
ppr Id
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
(DFuns like $fOrdList) lest we accidentally build a recursive
dictionary. See Note [Avoiding loops (DFuns)].

The basic strategy of Note [Avoiding loops (DFuns)] 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_imports 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
have 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.


************************************************************************
*                                                                      *
\subsubsection{@specExpr@: the main function}
*                                                                      *
************************************************************************
-}

data SpecEnv
  = SE { SpecEnv -> Subst
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!)

       , SpecEnv -> Module
se_module :: Module
       , SpecEnv -> RuleEnv
se_rules  :: RuleEnv  -- From the home package and this module
       , SpecEnv -> DynFlags
se_dflags :: DynFlags
     }

instance Outputable SpecEnv where
  ppr :: SpecEnv -> SDoc
ppr (SE { se_subst :: SpecEnv -> Subst
se_subst = Subst
subst })
    = String -> SDoc
forall doc. IsLine doc => String -> doc
text String
"SE" SDoc -> SDoc -> SDoc
forall doc. IsLine doc => doc -> doc -> doc
<+> SDoc -> SDoc
forall doc. IsLine doc => doc -> doc
braces (String -> SDoc
forall doc. IsLine doc => String -> doc
text String
"subst =" SDoc -> SDoc -> SDoc
forall doc. IsLine doc => doc -> doc -> doc
<+> Subst -> SDoc
forall a. Outputable a => a -> SDoc
ppr Subst
subst)

specVar :: SpecEnv -> InId -> SpecM (OutExpr, UsageDetails)
specVar :: SpecEnv -> Id -> SpecM (OutExpr, UsageDetails)
specVar env :: SpecEnv
env@(SE { se_subst :: SpecEnv -> Subst
se_subst = Core.Subst InScopeSet
in_scope IdSubstEnv
ids TvSubstEnv
_ CvSubstEnv
_ }) Id
v
  | Bool -> Bool
not (Id -> Bool
isLocalId Id
v)                   = (OutExpr, UsageDetails) -> SpecM (OutExpr, UsageDetails)
forall a. a -> UniqSM a
forall (m :: * -> *) a. Monad m => a -> m a
return (Id -> OutExpr
forall b. Id -> Expr b
Var Id
v, UsageDetails
emptyUDs)
  | Just OutExpr
e  <- IdSubstEnv -> Id -> Maybe OutExpr
forall a. VarEnv a -> Id -> Maybe a
lookupVarEnv IdSubstEnv
ids       Id
v = SpecEnv -> OutExpr -> SpecM (OutExpr, UsageDetails)
specExpr (SpecEnv -> SpecEnv
zapSubst SpecEnv
env) OutExpr
e  -- Note (1)
  | Just Id
v' <- InScopeSet -> Id -> Maybe Id
lookupInScope InScopeSet
in_scope Id
v = (OutExpr, UsageDetails) -> SpecM (OutExpr, UsageDetails)
forall a. a -> UniqSM a
forall (m :: * -> *) a. Monad m => a -> m a
return (Id -> OutExpr
forall b. Id -> Expr b
Var Id
v', UsageDetails
emptyUDs)
  | Bool
otherwise = String -> SDoc -> SpecM (OutExpr, UsageDetails)
forall a. HasCallStack => String -> SDoc -> a
pprPanic String
"specVar" (Id -> SDoc
forall a. Outputable a => a -> SDoc
ppr Id
v SDoc -> SDoc -> SDoc
forall doc. IsDoc doc => doc -> doc -> doc
$$ InScopeSet -> SDoc
forall a. Outputable a => a -> SDoc
ppr InScopeSet
in_scope)
  -- c.f. GHC.Core.Subst.lookupIdSubst
  -- Note (1): we recurse so we do the lookupInScope thing on any Vars in e
  --           probably has little effect, but it's the right thing.
  --           We need zapSubst because `e` is an OutExpr

specExpr :: SpecEnv -> CoreExpr -> SpecM (CoreExpr, UsageDetails)

---------------- First the easy cases --------------------
specExpr :: SpecEnv -> OutExpr -> SpecM (OutExpr, UsageDetails)
specExpr SpecEnv
env (Var Id
v)       = SpecEnv -> Id -> SpecM (OutExpr, UsageDetails)
specVar SpecEnv
env Id
v
specExpr SpecEnv
env (Type Kind
ty)     = (OutExpr, UsageDetails) -> SpecM (OutExpr, UsageDetails)
forall a. a -> UniqSM a
forall (m :: * -> *) a. Monad m => a -> m a
return (Kind -> OutExpr
forall b. Kind -> Expr b
Type     (SpecEnv -> Kind -> Kind
substTy SpecEnv
env Kind
ty), UsageDetails
emptyUDs)
specExpr SpecEnv
env (Coercion Coercion
co) = (OutExpr, UsageDetails) -> SpecM (OutExpr, UsageDetails)
forall a. a -> UniqSM a
forall (m :: * -> *) a. Monad m => a -> m a
return (Coercion -> OutExpr
forall b. Coercion -> Expr b
Coercion (SpecEnv -> Coercion -> Coercion
substCo SpecEnv
env Coercion
co), UsageDetails
emptyUDs)
specExpr SpecEnv
_   (Lit Literal
lit)     = (OutExpr, UsageDetails) -> SpecM (OutExpr, UsageDetails)
forall a. a -> UniqSM a
forall (m :: * -> *) a. Monad m => a -> m a
return (Literal -> OutExpr
forall b. Literal -> Expr b
Lit Literal
lit,                   UsageDetails
emptyUDs)
specExpr SpecEnv
env (Cast OutExpr
e Coercion
co)
  = do { (e', uds) <- SpecEnv -> OutExpr -> SpecM (OutExpr, UsageDetails)
specExpr SpecEnv
env OutExpr
e
       ; return ((mkCast e' (substCo env co)), uds) }
specExpr SpecEnv
env (Tick CoreTickish
tickish OutExpr
body)
  = do { (body', uds) <- SpecEnv -> OutExpr -> SpecM (OutExpr, UsageDetails)
specExpr SpecEnv
env OutExpr
body
       ; return (Tick (specTickish env tickish) body', uds) }

---------------- Applications might generate a call instance --------------------
specExpr SpecEnv
env expr :: OutExpr
expr@(App {})
  = do { let (OutExpr
fun_in, [OutExpr]
args_in) = OutExpr -> (OutExpr, [OutExpr])
forall b. Expr b -> (Expr b, [Expr b])
collectArgs OutExpr
expr
       ; (args_out, uds_args) <- (OutExpr -> SpecM (OutExpr, UsageDetails))
-> [OutExpr] -> SpecM ([OutExpr], UsageDetails)
forall a b.
(a -> SpecM (b, UsageDetails)) -> [a] -> SpecM ([b], UsageDetails)
mapAndCombineSM (SpecEnv -> OutExpr -> SpecM (OutExpr, UsageDetails)
specExpr SpecEnv
env) [OutExpr]
args_in
       ; let env_args = SpecEnv
env SpecEnv -> FloatedDictBinds -> SpecEnv
`bringFloatedDictsIntoScope` UsageDetails -> FloatedDictBinds
ud_binds UsageDetails
uds_args
                -- Some dicts may have floated out of args_in;
                -- they should be in scope for fireRewriteRules (#21689)
             (fun_in', args_out') = fireRewriteRules env_args fun_in args_out
       ; (fun_out', uds_fun) <- specExpr env fun_in'
       ; let uds_call = SpecEnv -> OutExpr -> [OutExpr] -> UsageDetails
mkCallUDs SpecEnv
env OutExpr
fun_out' [OutExpr]
args_out'
       ; return (fun_out' `mkApps` args_out', uds_fun `thenUDs` uds_call `thenUDs` uds_args) }

---------------- Lambda/case require dumping of usage details --------------------
specExpr SpecEnv
env e :: OutExpr
e@(Lam {})
  = SpecEnv -> [Id] -> OutExpr -> SpecM (OutExpr, UsageDetails)
specLam SpecEnv
env' [Id]
bndrs' OutExpr
body
  where
    ([Id]
bndrs, OutExpr
body)  = OutExpr -> ([Id], OutExpr)
forall b. Expr b -> ([b], Expr b)
collectBinders OutExpr
e
    (SpecEnv
env', [Id]
bndrs') = SpecEnv -> [Id] -> (SpecEnv, [Id])
forall (f :: * -> *).
Traversable f =>
SpecEnv -> f Id -> (SpecEnv, f Id)
substBndrs SpecEnv
env [Id]
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 SpecEnv
env (Case OutExpr
scrut Id
case_bndr Kind
ty [Alt Id]
alts)
  = do { (scrut', scrut_uds) <- SpecEnv -> OutExpr -> SpecM (OutExpr, UsageDetails)
specExpr SpecEnv
env OutExpr
scrut
       ; (scrut'', case_bndr', alts', alts_uds)
             <- specCase env scrut' case_bndr alts
--       ; pprTrace "specExpr:case" (vcat
--            [ text "scrut" <+> ppr scrut, text "scrut'" <+> ppr scrut'
--            , text "case_bndr'" <+> ppr case_bndr'
--            , text "alts_uds" <+> ppr alts_uds
--            ])
       ; return (Case scrut'' case_bndr' (substTy env ty) alts'
                , scrut_uds `thenUDs` alts_uds) }

---------------- Finally, let is the interesting case --------------------
specExpr SpecEnv
env (Let InBind
bind OutExpr
body)
  = do { (binds', body', uds) <- TopLevelFlag
-> SpecEnv
-> InBind
-> (SpecEnv -> SpecM (OutExpr, UsageDetails))
-> SpecM (CoreProgram, OutExpr, UsageDetails)
forall body.
TopLevelFlag
-> SpecEnv
-> InBind
-> (SpecEnv -> SpecM (body, UsageDetails))
-> SpecM (CoreProgram, body, UsageDetails)
specBind TopLevelFlag
NotTopLevel SpecEnv
env InBind
bind ((SpecEnv -> SpecM (OutExpr, UsageDetails))
 -> SpecM (CoreProgram, OutExpr, UsageDetails))
-> (SpecEnv -> SpecM (OutExpr, UsageDetails))
-> SpecM (CoreProgram, OutExpr, UsageDetails)
forall a b. (a -> b) -> a -> b
$ \SpecEnv
body_env ->
                                 -- pprTrace "specExpr:let" (ppr (se_subst body_env) $$ ppr body) $
                                 SpecEnv -> OutExpr -> SpecM (OutExpr, UsageDetails)
specExpr SpecEnv
body_env OutExpr
body
         -- All done
       ; return (foldr Let body' binds', uds) }

-- See Note [Specialisation modulo dictionary selectors]
--     Note [ClassOp/DFun selection]
--     Note [Fire rules in the specialiser]
fireRewriteRules :: SpecEnv -> InExpr -> [OutExpr] -> (InExpr, [OutExpr])
fireRewriteRules :: SpecEnv -> OutExpr -> [OutExpr] -> (OutExpr, [OutExpr])
fireRewriteRules SpecEnv
env (Var Id
f) [OutExpr]
args
  | let rules :: [CoreRule]
rules = RuleEnv -> Id -> [CoreRule]
getRules (SpecEnv -> RuleEnv
se_rules SpecEnv
env) Id
f
  , Just (CoreRule
rule, OutExpr
expr) <- SpecEnv
-> Id
-> [OutExpr]
-> (Activation -> Bool)
-> [CoreRule]
-> Maybe (CoreRule, OutExpr)
specLookupRule SpecEnv
env Id
f [OutExpr]
args Activation -> Bool
activeInInitialPhase [CoreRule]
rules
  , let rest_args :: [OutExpr]
rest_args    = Int -> [OutExpr] -> [OutExpr]
forall a. Int -> [a] -> [a]
drop (CoreRule -> Int
ruleArity CoreRule
rule) [OutExpr]
args -- See Note [Extra args in the target]
        zapped_subst :: Subst
zapped_subst = Subst -> Subst
Core.zapSubst (SpecEnv -> Subst
se_subst SpecEnv
env)
        expr' :: OutExpr
expr'        = HasDebugCallStack => SimpleOpts -> Subst -> OutExpr -> OutExpr
SimpleOpts -> Subst -> OutExpr -> OutExpr
simpleOptExprWith SimpleOpts
defaultSimpleOpts Subst
zapped_subst OutExpr
expr
                       -- simplOptExpr needed because lookupRule returns
                       --   (\x y. rhs) arg1 arg2
  , (OutExpr
fun', [OutExpr]
args') <- OutExpr -> (OutExpr, [OutExpr])
forall b. Expr b -> (Expr b, [Expr b])
collectArgs OutExpr
expr'
  = SpecEnv -> OutExpr -> [OutExpr] -> (OutExpr, [OutExpr])
fireRewriteRules SpecEnv
env OutExpr
fun' ([OutExpr]
args'[OutExpr] -> [OutExpr] -> [OutExpr]
forall a. [a] -> [a] -> [a]
++[OutExpr]
rest_args)
fireRewriteRules SpecEnv
_ OutExpr
fun [OutExpr]
args = (OutExpr
fun, [OutExpr]
args)

--------------
specLam :: SpecEnv -> [OutBndr] -> InExpr -> SpecM (OutExpr, UsageDetails)
-- The binders have been substituted, but the body has not
specLam :: SpecEnv -> [Id] -> OutExpr -> SpecM (OutExpr, UsageDetails)
specLam SpecEnv
env [Id]
bndrs OutExpr
body
  | [Id] -> Bool
forall a. [a] -> Bool
forall (t :: * -> *) a. Foldable t => t a -> Bool
null [Id]
bndrs
  = SpecEnv -> OutExpr -> SpecM (OutExpr, UsageDetails)
specExpr SpecEnv
env OutExpr
body
  | Bool
otherwise
  = do { (body', uds) <- SpecEnv -> OutExpr -> SpecM (OutExpr, UsageDetails)
specExpr SpecEnv
env OutExpr
body
       ; let (free_uds, dumped_dbs) = dumpUDs bndrs uds
       ; return (mkLams bndrs (wrapDictBindsE dumped_dbs body'), free_uds) }

--------------
specTickish :: SpecEnv -> CoreTickish -> CoreTickish
specTickish :: SpecEnv -> CoreTickish -> CoreTickish
specTickish (SE { se_subst :: SpecEnv -> Subst
se_subst = Subst
subst }) CoreTickish
bp = Subst -> CoreTickish -> CoreTickish
substTickish Subst
subst CoreTickish
bp

--------------
specCase :: SpecEnv
         -> OutExpr             -- Scrutinee, already done
         -> InId -> [InAlt]
         -> SpecM ( OutExpr     -- New scrutinee
                  , OutId
                  , [OutAlt]
                  , UsageDetails)
specCase :: SpecEnv
-> OutExpr
-> Id
-> [Alt Id]
-> SpecM (OutExpr, Id, [Alt Id], UsageDetails)
specCase SpecEnv
env OutExpr
scrut' Id
case_bndr [Alt AltCon
con [Id]
args OutExpr
rhs]
  | -- See Note [Floating dictionaries out of cases]
    OutExpr -> Kind -> Bool
interestingDict OutExpr
scrut' (Id -> Kind
idType Id
case_bndr)
  , Bool -> Bool
not (Id -> Bool
isDeadBinder Id
case_bndr Bool -> Bool -> Bool
&& [Id] -> Bool
forall a. [a] -> Bool
forall (t :: * -> *) a. Foldable t => t a -> Bool
null [Id]
sc_args')
  = do { case_bndr_flt :| sc_args_flt <- (Id -> UniqSM Id) -> NonEmpty Id -> UniqSM (NonEmpty Id)
forall (t :: * -> *) (m :: * -> *) a b.
(Traversable t, Monad m) =>
(a -> m b) -> t a -> m (t b)
forall (m :: * -> *) a b.
Monad m =>
(a -> m b) -> NonEmpty a -> m (NonEmpty b)
mapM Id -> UniqSM Id
forall {m :: * -> *}. MonadUnique m => Id -> m Id
clone_me (Id
case_bndr' Id -> [Id] -> NonEmpty Id
forall a. a -> [a] -> NonEmpty a
:| [Id]
sc_args')

       ; let case_bndr_flt' = Id
case_bndr_flt Id -> OutExpr -> Id
`addDictUnfolding` OutExpr
scrut'
             scrut_bind     = InBind -> DictBind
mkDB (Id -> OutExpr -> InBind
forall b. b -> Expr b -> Bind b
NonRec Id
case_bndr_flt OutExpr
scrut')

             sc_args_flt' = (Id -> OutExpr -> Id) -> [Id] -> [OutExpr] -> [Id]
forall a b c. (a -> b -> c) -> [a] -> [b] -> [c]
zipWith Id -> OutExpr -> Id
addDictUnfolding [Id]
sc_args_flt [OutExpr]
sc_rhss
             sc_rhss      = [ OutExpr -> Id -> Kind -> [Alt Id] -> OutExpr
forall b. Expr b -> b -> Kind -> [Alt b] -> Expr b
Case (Id -> OutExpr
forall b. Id -> Expr b
Var Id
case_bndr_flt') Id
case_bndr' (Id -> Kind
idType Id
sc_arg')
                                   [AltCon -> [Id] -> OutExpr -> Alt Id
forall b. AltCon -> [b] -> Expr b -> Alt b
Alt AltCon
con [Id]
args' (Id -> OutExpr
forall b. Id -> Expr b
Var Id
sc_arg')]
                            | Id
sc_arg' <- [Id]
sc_args' ]
             cb_set       = Id -> VarSet
unitVarSet Id
case_bndr_flt'
             sc_binds     = [ DB { db_bind :: InBind
db_bind = Id -> OutExpr -> InBind
forall b. b -> Expr b -> Bind b
NonRec Id
sc_arg_flt OutExpr
sc_rhs, db_fvs :: VarSet
db_fvs  = VarSet
cb_set }
                            | (Id
sc_arg_flt, OutExpr
sc_rhs) <- [Id]
sc_args_flt' [Id] -> [OutExpr] -> [(Id, OutExpr)]
forall a b. [a] -> [b] -> [(a, b)]
`zip` [OutExpr]
sc_rhss ]

             flt_binds    = DictBind
scrut_bind DictBind -> [DictBind] -> [DictBind]
forall a. a -> [a] -> [a]
: [DictBind]
sc_binds

             -- Extend the substitution for RHS to map the *original* binders
             -- to their floated versions.
             mb_sc_flts :: [Maybe DictId]
             mb_sc_flts = (Id -> Maybe Id) -> [Id] -> [Maybe Id]
forall a b. (a -> b) -> [a] -> [b]
map (VarEnv Id -> Id -> Maybe Id
forall a. VarEnv a -> Id -> Maybe a
lookupVarEnv VarEnv Id
clone_env) [Id]
args'
             clone_env  = [Id] -> [Id] -> VarEnv Id
forall a. [Id] -> [a] -> VarEnv a
zipVarEnv [Id]
sc_args' [Id]
sc_args_flt'

             subst_prs  = (Id
case_bndr, Id -> OutExpr
forall b. Id -> Expr b
Var Id
case_bndr_flt)
                        (Id, OutExpr) -> [(Id, OutExpr)] -> [(Id, OutExpr)]
forall a. a -> [a] -> [a]
: [ (Id
arg, Id -> OutExpr
forall b. Id -> Expr b
Var Id
sc_flt)
                          | (Id
arg, Just Id
sc_flt) <- [Id]
args [Id] -> [Maybe Id] -> [(Id, Maybe Id)]
forall a b. [a] -> [b] -> [(a, b)]
`zip` [Maybe Id]
mb_sc_flts ]
             subst'   = SpecEnv -> Subst
se_subst SpecEnv
env_rhs
                        Subst -> [Id] -> Subst
`Core.extendSubstInScopeList` (Id
case_bndr_flt' Id -> [Id] -> [Id]
forall a. a -> [a] -> [a]
: [Id]
sc_args_flt')
                        Subst -> [(Id, OutExpr)] -> Subst
`Core.extendIdSubstList`      [(Id, OutExpr)]
subst_prs
             env_rhs' = SpecEnv
env_rhs { se_subst = subst' }

       ; (rhs', rhs_uds)   <- specExpr env_rhs' rhs
       ; let (free_uds, dumped_dbs) = dumpUDs (case_bndr':args') rhs_uds
             all_uds = [DictBind]
flt_binds [DictBind] -> UsageDetails -> UsageDetails
`consDictBinds` UsageDetails
free_uds
             alt'    = AltCon -> [Id] -> OutExpr -> Alt Id
forall b. AltCon -> [b] -> Expr b -> Alt b
Alt AltCon
con [Id]
args' (OrdList DictBind -> OutExpr -> OutExpr
wrapDictBindsE OrdList DictBind
dumped_dbs OutExpr
rhs')
--       ; pprTrace "specCase" (ppr case_bndr $$ ppr scrut_bind) $
       ; return (Var case_bndr_flt, case_bndr', [alt'], all_uds) }
  where
    (SpecEnv
env_rhs, (Id
case_bndr':|[Id]
args')) = SpecEnv -> NonEmpty Id -> (SpecEnv, NonEmpty Id)
forall (f :: * -> *).
Traversable f =>
SpecEnv -> f Id -> (SpecEnv, f Id)
substBndrs SpecEnv
env (Id
case_bndrId -> [Id] -> NonEmpty Id
forall a. a -> [a] -> NonEmpty a
:|[Id]
args)
    sc_args' :: [Id]
sc_args' = (Id -> Bool) -> [Id] -> [Id]
forall a. (a -> Bool) -> [a] -> [a]
filter Id -> Bool
is_flt_sc_arg [Id]
args'

    clone_me :: Id -> m Id
clone_me Id
bndr = do { uniq <- m Unique
forall (m :: * -> *). MonadUnique m => m Unique
getUniqueM
                       ; return (mkUserLocalOrCoVar occ uniq wght ty loc) }
       where
         name :: Name
name = Id -> Name
idName Id
bndr
         wght :: Kind
wght = HasDebugCallStack => Id -> Kind
Id -> Kind
idMult Id
bndr
         ty :: Kind
ty   = Id -> Kind
idType Id
bndr
         occ :: OccName
occ  = Name -> OccName
nameOccName Name
name
         loc :: SrcSpan
loc  = Name -> SrcSpan
forall a. NamedThing a => a -> SrcSpan
getSrcSpan Name
name

    arg_set :: VarSet
arg_set = [Id] -> VarSet
mkVarSet [Id]
args'
    is_flt_sc_arg :: Id -> Bool
is_flt_sc_arg Id
var =  Id -> Bool
isId Id
var
                      Bool -> Bool -> Bool
&& Bool -> Bool
not (Id -> Bool
isDeadBinder Id
var)
                      Bool -> Bool -> Bool
&& Kind -> Bool
isDictTy Kind
var_ty
                      Bool -> Bool -> Bool
&& Kind -> VarSet
tyCoVarsOfType Kind
var_ty VarSet -> VarSet -> Bool
`disjointVarSet` VarSet
arg_set
       where
         var_ty :: Kind
var_ty = Id -> Kind
idType Id
var


specCase SpecEnv
env OutExpr
scrut Id
case_bndr [Alt Id]
alts
  = do { (alts', uds_alts) <- (Alt Id -> SpecM (Alt Id, UsageDetails))
-> [Alt Id] -> SpecM ([Alt Id], UsageDetails)
forall a b.
(a -> SpecM (b, UsageDetails)) -> [a] -> SpecM ([b], UsageDetails)
mapAndCombineSM Alt Id -> SpecM (Alt Id, UsageDetails)
spec_alt [Alt Id]
alts
       ; return (scrut, case_bndr', alts', uds_alts) }
  where
    (SpecEnv
env_alt, Id
case_bndr') = SpecEnv -> Id -> (SpecEnv, Id)
substBndr SpecEnv
env Id
case_bndr
    spec_alt :: Alt Id -> SpecM (Alt Id, UsageDetails)
spec_alt (Alt AltCon
con [Id]
args OutExpr
rhs)
      = do { (rhs', uds) <- SpecEnv -> OutExpr -> SpecM (OutExpr, UsageDetails)
specExpr SpecEnv
env_rhs OutExpr
rhs
           ; let (free_uds, dumped_dbs) = dumpUDs (case_bndr' : args') uds
--           ; unless (isNilOL dumped_dbs) $
--             pprTrace "specAlt" (vcat
--                 [text "case_bndr', args" <+> (ppr case_bndr' $$ ppr args)
--                 ,text "dumped" <+> ppr dumped_dbs ]) return ()
           ; return (Alt con args' (wrapDictBindsE dumped_dbs rhs'), free_uds) }
        where
          (SpecEnv
env_rhs, [Id]
args') = SpecEnv -> [Id] -> (SpecEnv, [Id])
forall (f :: * -> *).
Traversable f =>
SpecEnv -> f Id -> (SpecEnv, f Id)
substBndrs SpecEnv
env_alt [Id]
args

{- Note [Fire rules in the specialiser]
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Consider this (#21851)

    module A where
      f :: Num b => b -> (b, b)
      f x = (x + 1, snd (f x))
      {-# SPECIALIZE f :: Int -> (Int, Int) #-}

    module B (g') where
      import A

      g :: Num a => a -> a
      g x = fst (f x)
      {-# NOINLINE[99] g #-}

      h :: Int -> Int
      h = g

Note that `f` has the CPR property, and so will worker/wrapper.

The call to `g` in `h` will make us specialise `g @Int`. And the specialised
version of `g` will contain the call `f @Int`; but in the subsequent run of
the Simplifier, there will be a competition between:
  * The user-supplied SPECIALISE rule for `f`
  * The inlining of the wrapper for `f`
In fact, the latter wins -- see Note [tryRules: plan (BEFORE)]
GHC.Core.Opt.Simplify.Iteration.  However, it a bit fragile.

Moreover consider (test T21851_2):

    module A
      f :: (Ord a, Show b) => a -> b -> blah
      {-# RULE forall b. f @Int @b = wombat #-}

      wombat :: Show b => Int -> b -> blah
      wombat = blah

    module B
      import A
      g :: forall a. Ord a => blah
      g @a = ...g...f @a @Char....

      h = ....g @Int....

Now, in module B, GHC will specialise `g @Int`, which will lead to a
call `f @Int @Char`.  If we immediately (in the specialiser) rewrite
that to `womabat @Char`, we have a chance to specialise `wombat`.

Conclusion: it's treat if the Specialiser fires RULEs itself.
It's not hard to achieve: see `fireRewriteRules`. The only tricky bit is
making sure that we have a reasonably up to date EPS rule base. Currently
we load it up just once, in `initRuleEnv`, called at the beginning of
`specProgram`.

NB: you might wonder if running rules in the specialiser (this Note) renders
Note [tryRules: plan (BEFORE)] in the Simplifier (partly) redundant.  That is,
if we run rules in the specialiser, does it matter if we make rules "win" over
inlining in the Simplifier?  Yes, it does!  See the discussion in #21851.

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
*                                                                      *
************************************************************************
-}

bringFloatedDictsIntoScope :: SpecEnv -> FloatedDictBinds -> SpecEnv
bringFloatedDictsIntoScope :: SpecEnv -> FloatedDictBinds -> SpecEnv
bringFloatedDictsIntoScope SpecEnv
env (FDB { fdb_bndrs :: FloatedDictBinds -> VarSet
fdb_bndrs = VarSet
dx_bndrs })
  = -- pprTrace "brought into scope" (ppr dx_bndrs) $
    SpecEnv
env {se_subst=subst'}
  where
   subst' :: Subst
subst' = SpecEnv -> Subst
se_subst SpecEnv
env Subst -> VarSet -> Subst
`Core.extendSubstInScopeSet` VarSet
dx_bndrs

specBind :: TopLevelFlag
         -> SpecEnv    -- At top-level only, this env already has the
                       -- top level binders in scope
         -> InBind
         -> (SpecEnv -> SpecM (body, UsageDetails))    -- Process the body
         -> SpecM ( [OutBind]           -- New bindings
                  , body                -- Body
                  , UsageDetails)       -- And info to pass upstream

-- Returned UsageDetails:
--    No calls for binders of this bind
specBind :: forall body.
TopLevelFlag
-> SpecEnv
-> InBind
-> (SpecEnv -> SpecM (body, UsageDetails))
-> SpecM (CoreProgram, body, UsageDetails)
specBind TopLevelFlag
top_lvl SpecEnv
env (NonRec Id
fn OutExpr
rhs) SpecEnv -> SpecM (body, UsageDetails)
do_body
  = do { (rhs', rhs_uds) <- SpecEnv -> OutExpr -> SpecM (OutExpr, UsageDetails)
specExpr SpecEnv
env OutExpr
rhs

       ; (body_env1, fn1) <- case top_lvl of
                               TopLevelFlag
TopLevel    -> (SpecEnv, Id) -> UniqSM (SpecEnv, Id)
forall a. a -> UniqSM a
forall (m :: * -> *) a. Monad m => a -> m a
return (SpecEnv
env, Id
fn)
                               TopLevelFlag
NotTopLevel -> SpecEnv -> Id -> UniqSM (SpecEnv, Id)
cloneBndrSM SpecEnv
env Id
fn

       ; let fn2 | Unfolding -> Bool
isStableUnfolding (Id -> Unfolding
idUnfolding Id
fn1) = Id
fn1
                 | Bool
otherwise = Id
fn1 Id -> Unfolding -> Id
`setIdUnfolding` UnfoldingOpts -> OutExpr -> Unfolding
mkSimpleUnfolding UnfoldingOpts
defaultUnfoldingOpts OutExpr
rhs'
             -- Update the unfolding with the perhaps-simpler or more specialised rhs'
             -- This is important: see Note [Update unfolding after specialisation]
             -- And in any case cloneBndrSM discards non-Stable unfoldings

             fn3 = Id -> Id
floatifyIdDemandInfo Id
fn2
             -- We zap the demand info because the binding may float,
             -- which would invalidate the demand info (see #17810 for example).
             -- Destroying demand info is not terrible; specialisation is
             -- always followed soon by demand analysis.
             -- See Note [Floatifying demand info when floating] in GHC.Core.Opt.SetLevels

             body_env2 = SpecEnv
body_env1 SpecEnv -> FloatedDictBinds -> SpecEnv
`bringFloatedDictsIntoScope` UsageDetails -> FloatedDictBinds
ud_binds UsageDetails
rhs_uds
                                   SpecEnv -> Id -> SpecEnv
`extendInScope` Id
fn3
                                   -- bringFloatedDictsIntoScope: see #23567

       ; (body', body_uds) <- do_body body_env2

       ; (fn4, spec_defns, body_uds1) <- specDefn env body_uds fn3 rhs

       ; let (free_uds, dump_dbs, float_all) = dumpBindUDs [fn4] body_uds1
             all_free_uds                    = UsageDetails
free_uds UsageDetails -> UsageDetails -> UsageDetails
`thenUDs` UsageDetails
rhs_uds

             pairs = [(Id, OutExpr)]
spec_defns [(Id, OutExpr)] -> [(Id, OutExpr)] -> [(Id, OutExpr)]
forall a. [a] -> [a] -> [a]
++ [(Id
fn4, OutExpr
rhs')]
                        -- fn4 mentions the spec_defns in its rules,
                        -- so put the latter first

             final_binds :: [DictBind]
             -- See Note [From non-recursive to recursive]
             final_binds | Bool -> Bool
not (OrdList DictBind -> Bool
forall a. OrdList a -> Bool
isNilOL OrdList DictBind
dump_dbs)
                         , Bool -> Bool
not ([(Id, OutExpr)] -> Bool
forall a. [a] -> Bool
forall (t :: * -> *) a. Foldable t => t a -> Bool
null [(Id, OutExpr)]
spec_defns)
                         = [[(Id, OutExpr)] -> OrdList DictBind -> DictBind
recWithDumpedDicts [(Id, OutExpr)]
pairs OrdList DictBind
dump_dbs]
                         | Bool
otherwise
                         = [InBind -> DictBind
mkDB (InBind -> DictBind) -> InBind -> DictBind
forall a b. (a -> b) -> a -> b
$ Id -> OutExpr -> InBind
forall b. b -> Expr b -> Bind b
NonRec Id
b OutExpr
r | (Id
b,OutExpr
r) <- [(Id, OutExpr)]
pairs]
                           [DictBind] -> [DictBind] -> [DictBind]
forall a. [a] -> [a] -> [a]
++ OrdList DictBind -> [DictBind]
forall a. OrdList a -> [a]
fromOL OrdList DictBind
dump_dbs

             can_float_this_one = OutExpr -> Kind -> Bool
exprIsTopLevelBindable OutExpr
rhs (Id -> Kind
idType Id
fn)
             -- exprIsTopLevelBindable: see Note [Care with unlifted bindings]

       ; if float_all && can_float_this_one then
             -- Rather than discard the calls mentioning the bound variables
             -- we float this (dictionary) binding along with the others
              return ([], body', all_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, body', all_free_uds) }


specBind TopLevelFlag
top_lvl SpecEnv
env (Rec [(Id, OutExpr)]
pairs) SpecEnv -> SpecM (body, UsageDetails)
do_body
       -- Note [Specialising a recursive group]
  = do { let ([Id]
bndrs,[OutExpr]
rhss) = [(Id, OutExpr)] -> ([Id], [OutExpr])
forall (f :: * -> *) a b. Functor f => f (a, b) -> (f a, f b)
unzip [(Id, OutExpr)]
pairs

       ; (rec_env, bndrs1) <- case TopLevelFlag
top_lvl of
                                 TopLevelFlag
TopLevel    -> (SpecEnv, [Id]) -> UniqSM (SpecEnv, [Id])
forall a. a -> UniqSM a
forall (m :: * -> *) a. Monad m => a -> m a
return (SpecEnv
env, [Id]
bndrs)
                                 TopLevelFlag
NotTopLevel -> SpecEnv -> [Id] -> UniqSM (SpecEnv, [Id])
cloneRecBndrsSM SpecEnv
env [Id]
bndrs

       ; (rhss', rhs_uds)  <- mapAndCombineSM (specExpr rec_env) rhss
       ; (body', body_uds) <- do_body rec_env

       ; let scope_uds = UsageDetails
body_uds UsageDetails -> UsageDetails -> UsageDetails
`thenUDs` UsageDetails
rhs_uds
                       -- Includes binds and calls arising from rhss

       ; (bndrs2, spec_defns2, uds2) <- specDefns rec_env scope_uds (bndrs1 `zip` rhss)
         -- bndrs2 is like bndrs1, but with RULES added

       ; (bndrs3, spec_defns3, uds3)
             <- if null spec_defns2  -- Common case: no specialisation
                then return (bndrs2, [], uds2)
                else do {            -- Specialisation occurred; do it again
                          (bndrs3, spec_defns3, uds3)
                              <- specDefns rec_env uds2 (bndrs2 `zip` rhss)
                        ; return (bndrs3, spec_defns3 ++ spec_defns2, uds3) }

       ; let (final_uds, dumped_dbs, float_all) = dumpBindUDs bndrs1 uds3
             final_bind = [(Id, OutExpr)] -> OrdList DictBind -> DictBind
recWithDumpedDicts ([(Id, OutExpr)]
spec_defns3 [(Id, OutExpr)] -> [(Id, OutExpr)] -> [(Id, OutExpr)]
forall a. [a] -> [a] -> [a]
++ [Id] -> [OutExpr] -> [(Id, OutExpr)]
forall a b. [a] -> [b] -> [(a, b)]
zip [Id]
bndrs3 [OutExpr]
rhss')
                                             OrdList DictBind
dumped_dbs

       ; if float_all then
              return ([], body', final_uds `snocDictBind` final_bind)
         else
              return ([db_bind final_bind], body', 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 :: SpecEnv
-> UsageDetails
-> [(Id, OutExpr)]
-> SpecM ([Id], [(Id, OutExpr)], UsageDetails)
specDefns SpecEnv
_env UsageDetails
uds []
  = ([Id], [(Id, OutExpr)], UsageDetails)
-> SpecM ([Id], [(Id, OutExpr)], UsageDetails)
forall a. a -> UniqSM a
forall (m :: * -> *) a. Monad m => a -> m a
return ([], [], UsageDetails
uds)
specDefns SpecEnv
env UsageDetails
uds ((Id
bndr,OutExpr
rhs):[(Id, OutExpr)]
pairs)
  = do { (bndrs1, spec_defns1, uds1) <- SpecEnv
-> UsageDetails
-> [(Id, OutExpr)]
-> SpecM ([Id], [(Id, OutExpr)], UsageDetails)
specDefns SpecEnv
env UsageDetails
uds  [(Id, OutExpr)]
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 :: SpecEnv
-> UsageDetails
-> Id
-> OutExpr
-> SpecM (Id, [(Id, OutExpr)], UsageDetails)
specDefn SpecEnv
env UsageDetails
body_uds Id
fn OutExpr
rhs
  = do { let (UsageDetails
body_uds_without_me, [CallInfo]
calls_for_me) = Id -> UsageDetails -> (UsageDetails, [CallInfo])
callsForMe Id
fn UsageDetails
body_uds
             rules_for_me :: [CoreRule]
rules_for_me = Id -> [CoreRule]
idCoreRules Id
fn
             -- Bring into scope the binders from the floated dicts
             env_w_dict_bndrs :: SpecEnv
env_w_dict_bndrs = SpecEnv -> FloatedDictBinds -> SpecEnv
bringFloatedDictsIntoScope SpecEnv
env (UsageDetails -> FloatedDictBinds
ud_binds UsageDetails
body_uds)

       ; (rules, spec_defns, spec_uds) <- Bool
-> SpecEnv
-> [CoreRule]
-> [CallInfo]
-> Id
-> OutExpr
-> SpecM ([CoreRule], [(Id, OutExpr)], UsageDetails)
specCalls Bool
False SpecEnv
env_w_dict_bndrs
                                                    [CoreRule]
rules_for_me [CallInfo]
calls_for_me Id
fn OutExpr
rhs

       ; return ( fn `addIdSpecialisations` rules
                , spec_defns
                , body_uds_without_me `thenUDs` spec_uds) }
                -- It's important that the `thenUDs` 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 `alreadyCovered`

type SpecInfo = ( [CoreRule]       -- Specialisation rules
                , [(Id,CoreExpr)]  -- Specialised definition
                , UsageDetails )   -- Usage details from specialised RHSs

specCalls :: Bool
-> SpecEnv
-> [CoreRule]
-> [CallInfo]
-> Id
-> OutExpr
-> SpecM ([CoreRule], [(Id, OutExpr)], UsageDetails)
specCalls Bool
spec_imp SpecEnv
env [CoreRule]
existing_rules [CallInfo]
calls_for_me Id
fn OutExpr
rhs
        -- The first case is the interesting one
  |  [CallInfo] -> Bool
forall (f :: * -> *) a. Foldable f => f a -> Bool
notNull [CallInfo]
calls_for_me               -- And there are some calls to specialise
  Bool -> Bool -> Bool
&& Bool -> Bool
not (Activation -> Bool
isNeverActive (Id -> Activation
idInlineActivation Id
fn))
        -- Don't specialise NOINLINE things
        -- See Note [Auto-specialisation and RULES]
        --
        -- Don't specialise OPAQUE things, see Note [OPAQUE pragma].
        -- Since OPAQUE things are always never-active (see
        -- GHC.Parser.PostProcess.mkOpaquePragma) this guard never fires for
        -- OPAQUE things.

--   && not (certainlyWillInline (idUnfolding fn))      -- And it's not small
--      See Note [Inline specialisations] for why we do not
--      switch off specialisation for inline functions

  = -- pprTrace "specCalls: some" (vcat
    --   [ text "function" <+> ppr fn
    --    , text "calls:" <+> ppr calls_for_me
    --    , text "subst" <+> ppr (se_subst env) ]) $
    (([CoreRule], [(Id, OutExpr)], UsageDetails)
 -> CallInfo -> SpecM ([CoreRule], [(Id, OutExpr)], UsageDetails))
-> ([CoreRule], [(Id, OutExpr)], UsageDetails)
-> [CallInfo]
-> SpecM ([CoreRule], [(Id, OutExpr)], UsageDetails)
forall (t :: * -> *) (m :: * -> *) b a.
(Foldable t, Monad m) =>
(b -> a -> m b) -> b -> t a -> m b
foldlM ([CoreRule], [(Id, OutExpr)], UsageDetails)
-> CallInfo -> SpecM ([CoreRule], [(Id, OutExpr)], UsageDetails)
spec_call ([], [], UsageDetails
emptyUDs) [CallInfo]
calls_for_me

  | Bool
otherwise   -- No calls or RHS doesn't fit our preconceptions
  = Bool
-> String
-> SDoc
-> SpecM ([CoreRule], [(Id, OutExpr)], UsageDetails)
-> SpecM ([CoreRule], [(Id, OutExpr)], UsageDetails)
forall a. HasCallStack => Bool -> String -> SDoc -> a -> a
warnPprTrace (Bool -> Bool
not (OutExpr -> Bool
exprIsTrivial OutExpr
rhs) Bool -> Bool -> Bool
&& [CallInfo] -> Bool
forall (f :: * -> *) a. Foldable f => f a -> Bool
notNull [CallInfo]
calls_for_me)
          String
"Missed specialisation opportunity for" (Id -> SDoc
forall a. Outputable a => a -> SDoc
ppr Id
fn SDoc -> SDoc -> SDoc
forall doc. IsDoc doc => doc -> doc -> doc
$$ SDoc
trace_doc) (SpecM ([CoreRule], [(Id, OutExpr)], UsageDetails)
 -> SpecM ([CoreRule], [(Id, OutExpr)], UsageDetails))
-> SpecM ([CoreRule], [(Id, OutExpr)], UsageDetails)
-> SpecM ([CoreRule], [(Id, OutExpr)], UsageDetails)
forall a b. (a -> b) -> a -> b
$
          -- Note [Specialisation shape]
    -- pprTrace "specCalls: none" (ppr fn <+> ppr calls_for_me) $
    ([CoreRule], [(Id, OutExpr)], UsageDetails)
-> SpecM ([CoreRule], [(Id, OutExpr)], UsageDetails)
forall a. a -> UniqSM a
forall (m :: * -> *) a. Monad m => a -> m a
return ([], [], UsageDetails
emptyUDs)
  where
    trace_doc :: SDoc
trace_doc = [SDoc] -> SDoc
forall doc. IsLine doc => [doc] -> doc
sep [ [Id] -> SDoc
forall a. Outputable a => a -> SDoc
ppr [Id]
rhs_bndrs, Activation -> SDoc
forall a. Outputable a => a -> SDoc
ppr (Id -> Activation
idInlineActivation Id
fn) ]

    fn_type :: Kind
fn_type   = Id -> Kind
idType Id
fn
    fn_arity :: Int
fn_arity  = Id -> Int
idArity Id
fn
    fn_unf :: Unfolding
fn_unf    = Id -> Unfolding
realIdUnfolding Id
fn  -- Ignore loop-breaker-ness here
    inl_prag :: InlinePragma
inl_prag  = Id -> InlinePragma
idInlinePragma Id
fn
    inl_act :: Activation
inl_act   = InlinePragma -> Activation
inlinePragmaActivation InlinePragma
inl_prag
    is_active :: Activation -> Bool
is_active = CompilerPhase -> Activation -> Bool
isActive (Activation -> CompilerPhase
beginPhase Activation
inl_act) :: Activation -> Bool
         -- is_active: inl_act is the activation we are going to put in the new
         --   SPEC rule; so we want to see if it is covered by another rule with
         --   that same activation.
    is_local :: Bool
is_local  = Id -> Bool
isLocalId Id
fn
    is_dfun :: Bool
is_dfun   = Id -> Bool
isDFunId Id
fn
    dflags :: DynFlags
dflags    = SpecEnv -> DynFlags
se_dflags SpecEnv
env
    this_mod :: Module
this_mod  = SpecEnv -> Module
se_module SpecEnv
env
        -- Figure out whether the function has an INLINE pragma
        -- See Note [Inline specialisations]

    ([Id]
rhs_bndrs, OutExpr
rhs_body) = OutExpr -> ([Id], OutExpr)
collectBindersPushingCo OutExpr
rhs
                            -- See Note [Account for casts in binding]

    ----------------------------------------------------------
        -- Specialise to one particular call pattern
    spec_call :: SpecInfo                         -- Accumulating parameter
              -> CallInfo                         -- Call instance
              -> SpecM SpecInfo
    spec_call :: ([CoreRule], [(Id, OutExpr)], UsageDetails)
-> CallInfo -> SpecM ([CoreRule], [(Id, OutExpr)], UsageDetails)
spec_call spec_acc :: ([CoreRule], [(Id, OutExpr)], UsageDetails)
spec_acc@([CoreRule]
rules_acc, [(Id, OutExpr)]
pairs_acc, UsageDetails
uds_acc) _ci :: CallInfo
_ci@(CI { ci_key :: CallInfo -> [SpecArg]
ci_key = [SpecArg]
call_args })
      = -- See Note [Specialising Calls]
        do { let all_call_args :: [SpecArg]
all_call_args | Bool
is_dfun   = [SpecArg]
saturating_call_args -- See Note [Specialising DFuns]
                               | Bool
otherwise = [SpecArg]
call_args
                 saturating_call_args :: [SpecArg]
saturating_call_args = [SpecArg]
call_args [SpecArg] -> [SpecArg] -> [SpecArg]
forall a. [a] -> [a] -> [a]
++ (Id -> SpecArg) -> [Id] -> [SpecArg]
forall a b. (a -> b) -> [a] -> [b]
map Id -> SpecArg
mk_extra_dfun_arg ([SpecArg] -> [Id] -> [Id]
forall b a. [b] -> [a] -> [a]
dropList [SpecArg]
call_args [Id]
rhs_bndrs)
                 mk_extra_dfun_arg :: Id -> SpecArg
mk_extra_dfun_arg Id
bndr | Id -> Bool
isTyVar Id
bndr = SpecArg
UnspecType
                                        | Bool
otherwise = SpecArg
UnspecArg

           ; ( useful, rhs_env2, leftover_bndrs
             , rule_bndrs, rule_lhs_args
             , spec_bndrs1, dx_binds, spec_args) <- SpecEnv
-> [Id]
-> [SpecArg]
-> SpecM
     (Bool, SpecEnv, [Id], [Id], [OutExpr], [Id], [DictBind], [OutExpr])
specHeader SpecEnv
env [Id]
rhs_bndrs [SpecArg]
all_call_args

--           ; pprTrace "spec_call" (vcat
--                [ text "fun:       "  <+> ppr fn
--                , text "call info: "  <+> ppr _ci
--                , text "useful:    "  <+> ppr useful
--                , text "rule_bndrs:"  <+> ppr rule_bndrs
--                , text "lhs_args:  "  <+> ppr rule_lhs_args
--                , text "spec_bndrs1:" <+> ppr spec_bndrs1
--                , text "leftover_bndrs:" <+> pprIds leftover_bndrs
--                , text "spec_args: "  <+> ppr spec_args
--                , text "dx_binds:  "  <+> ppr dx_binds
--                , text "rhs_bndrs"     <+> ppr rhs_bndrs
--                , text "rhs_body"     <+> ppr rhs_body
--                , text "rhs_env2:  "  <+> ppr (se_subst rhs_env2)
--                , ppr dx_binds ]) $
--             return ()

           ; let all_rules = [CoreRule]
rules_acc [CoreRule] -> [CoreRule] -> [CoreRule]
forall a. [a] -> [a] -> [a]
++ [CoreRule]
existing_rules
                 -- all_rules: we look both in the rules_acc (generated by this invocation
                 --   of specCalls), and in existing_rules (passed in to specCalls)
           ; if not useful  -- No useful specialisation
                || alreadyCovered rhs_env2 rule_bndrs fn rule_lhs_args is_active all_rules
                   -- See (SC1) in Note [Specialisations already covered]
             then return spec_acc
             else
        do { -- Run the specialiser on the specialised RHS
             -- The "1" suffix is before we maybe add the void arg
           ; (rhs_body', rhs_uds) <- specExpr rhs_env2 rhs_body
                -- Add the { d1' = dx1; d2' = dx2 } usage stuff
                -- to the rhs_uds; see Note [Specialising Calls]
           ; let rhs_uds_w_dx   = [DictBind]
dx_binds [DictBind] -> UsageDetails -> UsageDetails
`consDictBinds` UsageDetails
rhs_uds
                 spec_rhs_bndrs = [Id]
spec_bndrs1 [Id] -> [Id] -> [Id]
forall a. [a] -> [a] -> [a]
++ [Id]
leftover_bndrs
                 (spec_uds, dumped_dbs) = dumpUDs spec_rhs_bndrs rhs_uds_w_dx
                 spec_rhs1 = [Id] -> OutExpr -> OutExpr
forall b. [b] -> Expr b -> Expr b
mkLams [Id]
spec_rhs_bndrs (OutExpr -> OutExpr) -> OutExpr -> OutExpr
forall a b. (a -> b) -> a -> b
$
                             OrdList DictBind -> OutExpr -> OutExpr
wrapDictBindsE OrdList DictBind
dumped_dbs OutExpr
rhs_body'

                 spec_fn_ty1 = HasDebugCallStack => OutExpr -> Kind
OutExpr -> Kind
exprType OutExpr
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.needsVoidWorkerArg
                 add_void_arg = HasDebugCallStack => Kind -> Bool
Kind -> Bool
isUnliftedType Kind
spec_fn_ty1 Bool -> Bool -> Bool
&& Bool -> Bool
not (Id -> Bool
isJoinId Id
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 = [OutExpr] -> Int
forall a. [a] -> Int
forall (t :: * -> *) a. Foldable t => t a -> Int
length [OutExpr]
rule_lhs_args Int -> Int -> Int
forall a. Num a => a -> a -> a
- [Id] -> Int
forall a. [a] -> Int
forall (t :: * -> *) a. Foldable t => t a -> Int
length [Id]
spec_bndrs

                 --------------------------------------
                 -- Add a suitable unfolding; see Note [Inline specialisations]
                 -- The wrap_unf_body applies the original unfolding to the specialised
                 -- arguments, not forgetting to wrap the dx_binds around the outside (#22358)
                 simpl_opts = DynFlags -> SimpleOpts
initSimpleOpts DynFlags
dflags
                 wrap_unf_body OutExpr
body = (DictBind -> OutExpr -> OutExpr)
-> OutExpr -> [DictBind] -> OutExpr
forall a b. (a -> b -> b) -> b -> [a] -> b
forall (t :: * -> *) a b.
Foldable t =>
(a -> b -> b) -> b -> t a -> b
foldr (InBind -> OutExpr -> OutExpr
forall b. Bind b -> Expr b -> Expr b
Let (InBind -> OutExpr -> OutExpr)
-> (DictBind -> InBind) -> DictBind -> OutExpr -> OutExpr
forall b c a. (b -> c) -> (a -> b) -> a -> c
. DictBind -> InBind
db_bind) (OutExpr
body OutExpr -> [OutExpr] -> OutExpr
forall b. Expr b -> [Expr b] -> Expr b
`mkApps` [OutExpr]
spec_args) [DictBind]
dx_binds
                 spec_unf = SimpleOpts
-> [Id]
-> (OutExpr -> OutExpr)
-> [OutExpr]
-> Unfolding
-> Unfolding
specUnfolding SimpleOpts
simpl_opts [Id]
spec_bndrs OutExpr -> OutExpr
wrap_unf_body
                                          [OutExpr]
rule_lhs_args Unfolding
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     = (OutExpr -> Bool) -> [OutExpr] -> Int
forall a. (a -> Bool) -> [a] -> Int
count OutExpr -> Bool
forall b. Expr b -> Bool
isValArg [OutExpr]
rule_lhs_args Int -> Int -> Int
forall a. Num a => a -> a -> a
- (Id -> Bool) -> [Id] -> Int
forall a. (a -> Bool) -> [a] -> Int
count Id -> Bool
isId [Id]
spec_bndrs

                 spec_inl_prag
                   | Bool -> Bool
not Bool
is_local     -- See Note [Specialising imported functions]
                   , OccInfo -> Bool
isStrongLoopBreaker (Id -> OccInfo
idOccInfo Id
fn) -- in GHC.Core.Opt.OccurAnal
                   = InlinePragma
neverInlinePragma
                   | Bool
otherwise
                   = InlinePragma
inl_prag

                 spec_fn_info
                   = IdInfo
vanillaIdInfo IdInfo -> Int -> IdInfo
`setArityInfo`      Int -> Int -> Int
forall a. Ord a => a -> a -> a
max Int
0 (Int
fn_arity Int -> Int -> Int
forall a. Num a => a -> a -> a
- Int
arity_decr)
                                   IdInfo -> InlinePragma -> IdInfo
`setInlinePragInfo` InlinePragma
spec_inl_prag
                                   IdInfo -> Unfolding -> IdInfo
`setUnfoldingInfo`  Unfolding
spec_unf

                 -- Compute the IdDetails of the specialise Id
                 -- See Note [Transfer IdDetails during specialisation]
                 spec_fn_details
                   = case Id -> IdDetails
idDetails Id
fn of
                       JoinId Int
join_arity Maybe [CbvMark]
_ -> Int -> Maybe [CbvMark] -> IdDetails
JoinId (Int
join_arity Int -> Int -> Int
forall a. Num a => a -> a -> a
- Int
join_arity_decr) Maybe [CbvMark]
forall a. Maybe a
Nothing
                       DFunId Bool
unary        -> Bool -> IdDetails
DFunId Bool
unary
                       IdDetails
_                   -> IdDetails
VanillaId

           ; spec_fn <- newSpecIdSM (idName fn) spec_fn_ty spec_fn_details spec_fn_info
           ; 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 | Bool
spec_imp  = -- Specialising imported fn
                                     String -> SDoc
forall doc. IsLine doc => String -> doc
text String
"SPEC/" SDoc -> SDoc -> SDoc
forall doc. IsLine doc => doc -> doc -> doc
<> Module -> SDoc
forall a. Outputable a => a -> SDoc
ppr Module
this_mod
                       | Bool
otherwise = -- Specialising local fn
                                     String -> SDoc
forall doc. IsLine doc => String -> doc
text String
"SPEC"

                spec_rule = DynFlags
-> Module
-> Bool
-> Activation
-> SDoc
-> Id
-> [Id]
-> [OutExpr]
-> OutExpr
-> CoreRule
mkSpecRule DynFlags
dflags Module
this_mod Bool
True Activation
inl_act
                                    SDoc
herald Id
fn [Id]
rule_bndrs [OutExpr]
rule_lhs_args
                                    (OutExpr -> [Id] -> OutExpr
forall b. Expr b -> [Id] -> Expr b
mkVarApps (Id -> OutExpr
forall b. Id -> Expr b
Var Id
spec_fn) [Id]
spec_bndrs)

                spec_f_w_arity = Id
spec_fn

                _rule_trace_doc = [SDoc] -> SDoc
forall doc. IsDoc doc => [doc] -> doc
vcat [ Id -> SDoc
forall a. Outputable a => a -> SDoc
ppr Id
fn SDoc -> SDoc -> SDoc
forall doc. IsLine doc => doc -> doc -> doc
<+> SDoc
dcolon SDoc -> SDoc -> SDoc
forall doc. IsLine doc => doc -> doc -> doc
<+> Kind -> SDoc
forall a. Outputable a => a -> SDoc
ppr Kind
fn_type
                                       , Id -> SDoc
forall a. Outputable a => a -> SDoc
ppr Id
spec_fn  SDoc -> SDoc -> SDoc
forall doc. IsLine doc => doc -> doc -> doc
<+> SDoc
dcolon SDoc -> SDoc -> SDoc
forall doc. IsLine doc => doc -> doc -> doc
<+> Kind -> SDoc
forall a. Outputable a => a -> SDoc
ppr Kind
spec_fn_ty
                                       , [Id] -> SDoc
forall a. Outputable a => a -> SDoc
ppr [Id]
rhs_bndrs, [SpecArg] -> SDoc
forall a. Outputable a => a -> SDoc
ppr [SpecArg]
call_args
                                       , CoreRule -> SDoc
forall a. Outputable a => a -> SDoc
ppr CoreRule
spec_rule
                                       , String -> SDoc
forall doc. IsLine doc => String -> doc
text String
"acc" SDoc -> SDoc -> SDoc
forall doc. IsLine doc => doc -> doc -> doc
<+> [CoreRule] -> SDoc
forall a. Outputable a => a -> SDoc
ppr [CoreRule]
rules_acc
                                       , String -> SDoc
forall doc. IsLine doc => String -> doc
text String
"existing" SDoc -> SDoc -> SDoc
forall doc. IsLine doc => doc -> doc -> doc
<+> [CoreRule] -> SDoc
forall a. Outputable a => a -> SDoc
ppr [CoreRule]
existing_rules
                                       ]

           ; -- pprTrace "spec_call: rule" _rule_trace_doc
             return ( spec_rule                  : rules_acc
                    , (spec_f_w_arity, spec_rhs) : pairs_acc
                    , spec_uds           `thenUDs` uds_acc
                    ) } }

alreadyCovered :: SpecEnv
               -> [Var] -> Id -> [CoreExpr]   -- LHS of possible new rule
               -> (Activation -> Bool)        -- Which rules are active
               -> [CoreRule] -> Bool
-- Note [Specialisations already covered] esp (SC2)
alreadyCovered :: SpecEnv
-> [Id]
-> Id
-> [OutExpr]
-> (Activation -> Bool)
-> [CoreRule]
-> Bool
alreadyCovered SpecEnv
env [Id]
bndrs Id
fn [OutExpr]
args Activation -> Bool
is_active [CoreRule]
rules
  = case SpecEnv
-> Id
-> [OutExpr]
-> (Activation -> Bool)
-> [CoreRule]
-> Maybe (CoreRule, OutExpr)
specLookupRule SpecEnv
env Id
fn [OutExpr]
args Activation -> Bool
is_active [CoreRule]
rules of
      Maybe (CoreRule, OutExpr)
Nothing             -> Bool
False
      Just (CoreRule
rule, OutExpr
_)
        | CoreRule -> Bool
isAutoRule CoreRule
rule -> -- Discard identical rules
                             -- We know that (fn args) is an instance of RULE
                             -- Check if RULE is an instance of (fn args)
                             InScopeSet -> [Id] -> [OutExpr] -> CoreRule -> Bool
ruleLhsIsMoreSpecific InScopeSet
in_scope [Id]
bndrs [OutExpr]
args CoreRule
rule
        | Bool
otherwise       -> Bool
True  -- User rules dominate
  where
    in_scope :: InScopeSet
in_scope = Subst -> InScopeSet
substInScopeSet (SpecEnv -> Subst
se_subst SpecEnv
env)

-- Convenience function for invoking lookupRule from Specialise
-- The SpecEnv's InScopeSet should include all the Vars in the [CoreExpr]
specLookupRule :: SpecEnv -> Id -> [CoreExpr]
               -> (Activation -> Bool)  -- Which rules are active
               -> [CoreRule] -> Maybe (CoreRule, CoreExpr)
specLookupRule :: SpecEnv
-> Id
-> [OutExpr]
-> (Activation -> Bool)
-> [CoreRule]
-> Maybe (CoreRule, OutExpr)
specLookupRule SpecEnv
env Id
fn [OutExpr]
args Activation -> Bool
is_active [CoreRule]
rules
  = RuleOpts
-> InScopeEnv
-> (Activation -> Bool)
-> Id
-> [OutExpr]
-> [CoreRule]
-> Maybe (CoreRule, OutExpr)
lookupRule RuleOpts
ropts InScopeEnv
in_scope_env Activation -> Bool
is_active Id
fn [OutExpr]
args [CoreRule]
rules
  where
    dflags :: DynFlags
dflags       = SpecEnv -> DynFlags
se_dflags SpecEnv
env
    in_scope :: InScopeSet
in_scope     = Subst -> InScopeSet
substInScopeSet (SpecEnv -> Subst
se_subst SpecEnv
env)
    in_scope_env :: InScopeEnv
in_scope_env = InScopeSet -> (Id -> Unfolding) -> InScopeEnv
ISE InScopeSet
in_scope ((Activation -> Bool) -> Id -> Unfolding
whenActiveUnfoldingFun Activation -> Bool
is_active)
    ropts :: RuleOpts
ropts        = DynFlags -> RuleOpts
initRuleOpts DynFlags
dflags

{- Note [Specialising DFuns]
~~~~~~~~~~~~~~~~~~~~~~~~~~~~
DFuns have a special sort of unfolding (DFunUnfolding), and it is
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 UnspecTypes/UnspecArgs,
so that 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.Make.specUnfolding.

Note [Transfer IdDetails during specialisation]
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
When specialising a function, `newSpecIdSM` comes up with a fresh Id the
specialised RHS will be bound to. It is critical that we get the `IdDetails` of
the specialised Id correct:

* JoinId: We want the specialised Id to be a join point, too.  But
  we have to carefully adjust the arity

* DFunId: It is crucial that we also make the new Id a DFunId.
  - First, because it obviously /is/ a DFun, having a DFunUnfolding and
    all that; see Note [Specialising DFuns]

  - Second, DFuns get very delicate special treatment in the demand analyser;
    see GHC.Core.Opt.DmdAnal.enterDFun.  If the specialised function isn't
    also a DFunId, this special treatment doesn't happen, so the demand
    analyser makes a too-strict DFun, and we get an infinite loop.  See Note
    [Do not strictify a DFun's parameter dictionaries] in GHC.Core.Opt.DmdAnal.
    #22549 describes the loop, and (lower down) a case where a /specialised/
    DFun caused a loop.

* WorkerLikeId: Introduced by WW, so after Specialise. Nevertheless, they come
  up when specialising imports. We must keep them as VanillaIds because WW
  will detect them as WorkerLikeIds again. That is, unless specialisation
  allows unboxing of all previous CBV args, in which case sticking to
  VanillaIds was the only correct choice to begin with.

* RecSelId, DataCon*Id, ClassOpId, PrimOpId, FCallId, CoVarId, TickBoxId:
  Never specialised.

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 [Care with unlifted bindings]
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Consider (#22998)
    f x = let x::ByteArray# = <some literal>
              n::Natural    = NB x
          in wombat @192827 (n |> co)
where
  co :: Natural ~ KnownNat 192827
  wombat :: forall (n:Nat). KnownNat n => blah

Left to itself, the specialiser would float the bindings for `x` and `n` to top
level, so we can specialise `wombat`.  But we can't have a top-level ByteArray#
(see Note [Core letrec invariant] in GHC.Core).  Boo.

This is pretty exotic, so we take a simple way out: in specBind (the NonRec
case) do not float the binding itself unless it satisfies exprIsTopLevelBindable.
This is conservative: maybe the RHS of `x` has a free var that would stop it
floating to top level anyway; but that is hard to spot (since we don't know what
the non-top-level in-scope binders are) and rare (since the binding must satisfy
Note [Core let-can-float invariant] in GHC.Core).


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.

Wrinkles

* Note 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.

* If the function has a stable unfolding, specHeader has to come up with
  arguments to pass to that stable unfolding, when building the stable
  unfolding of the specialised function: this is the last field in specHeader's
  big result tuple.

  The right thing to do is to produce a LitRubbish; it should rapidly
  disappear.  Rather like GHC.Core.Opt.WorkWrap.Utils.mk_absent_let.

Note [Specialisation modulo dictionary selectors]
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
In #19644, we discovered that the ClassOp/DFun rules from
Note [ClassOp/DFun selection] inhibit transitive specialisation.
Example, inspired by T17966:

  class C a where
    m :: Show b => a -> b -> String
    dummy :: a -> () -- Force a datatype dictionary representation

  instance C Int where
    m a b = show a ++ show b
    dummy _ = ()

  f :: (C a, Show b) => a -> b -> String
  f a b = m a b ++ "!"
  {-# INLINABLE[0] f #-}

  main = putStrLn (f (42::Int) (True::Bool))

Here, we specialise `f` at `Int` and `Bool`, giving

  $dC = $fCInt
  $dShow = GHC.Show.$fShowBool
  $sf (a::Int) (b::Bool) =
        ... (m @Int $dC @Bool $dShow a b) ...

Here `m` is just a DictSel, so there is (apparently) nothing to specialise!
However, the next Simplifier run will expose the rewritten instance method:

  ... $fCInt_$cm @Bool $fShowBool a b ...

where $fCInt_$cm is the instance method for `m` in `instance C Int`:

   $fCInt_$cm :: forall b. Show b => Int -> b -> String
   $fCInt_$cm b d x y = show @Int $dShowInt x ++ show @b d y

We want to specialise this! How? By doing the method-selection rewrite in
the Specialiser. Hence

1. In the App case of 'specExpr', try to apply the ClassOp/DFun rule on the
   head of the application, repeatedly, via 'fireRewriteRules'.
2. Attach an unfolding to freshly-bound dictionary ids such as `$dC` and
   `$dShow` in `bindAuxiliaryDict`, so that we can exploit the unfolding
   in 'fireRewriteRules' to do the ClassOp/DFun rewrite.

NB: Without (2), (1) would be pointless, because 'lookupRule' wouldn't be able
to look into the RHS of `$dC` to see the DFun.

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 (DFuns)]
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
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'

Note [Avoiding loops (non-DFuns)]
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
The whole Note [Avoiding loops (DFuns)] things applies only to DFuns.
It's important /not/ to apply filterCalls to non-DFuns. For 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.)
See test simplCore/should_compile/T19599a.

Another example is #19599, which looked like this:

   class (Show a, Enum a) => MyShow a where
      myShow :: a -> String

   myShow_impl :: MyShow a => a -> String

   foo :: Int -> String
   foo = myShow_impl @Int $fMyShowInt

   Rec { $fMyShowInt = MkMyShowD $fEnumInt $fShowInt $cmyShow
       ; $cmyShow = myShow_impl @Int $fMyShowInt }

Here, we really do want to specialise `myShow_impl @Int $fMyShowInt`.


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.  Wrinkles

(SC1) 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

(SC2) What about two specialisations where the second is an *instance*
   of the first?  It's a bit arbitrary, but here's what we do:
   * If the existing one is user-specified, via a SPECIALISE pragma, we
     suppress the further specialisation.
   * If the existing one is auto-generated, we generate a second RULE
     for the more specialised version.
   The latter is important because we don't want the accidental order
   of calls to determine what specialisations we generate.

(SC3) Annoyingly, we /also/ eliminate duplicates in `filterCalls`.
   See (MP3) in Note [Specialising polymorphic dictionaries]

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 inherited from the original function

  * InlineSpec: inherit from original function

  * Unfolding: transfer a StableUnfolding iff it is UnfWhen
               See GHC.Core.Unfold.Make.specUnfolding
               and its Note [Specialising unfoldings]

InlineSpec: 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.
-}

{- *********************************************************************
*                                                                      *
                   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 :: SpecArg -> SDoc
ppr (SpecType Kind
t) = String -> SDoc
forall doc. IsLine doc => String -> doc
text String
"SpecType" SDoc -> SDoc -> SDoc
forall doc. IsLine doc => doc -> doc -> doc
<+> Kind -> SDoc
forall a. Outputable a => a -> SDoc
ppr Kind
t
  ppr SpecArg
UnspecType   = String -> SDoc
forall doc. IsLine doc => String -> doc
text String
"UnspecType"
  ppr (SpecDict OutExpr
d) = String -> SDoc
forall doc. IsLine doc => String -> doc
text String
"SpecDict" SDoc -> SDoc -> SDoc
forall doc. IsLine doc => doc -> doc -> doc
<+> OutExpr -> SDoc
forall a. Outputable a => a -> SDoc
ppr OutExpr
d
  ppr SpecArg
UnspecArg    = String -> SDoc
forall doc. IsLine doc => String -> doc
text String
"UnspecArg"

specArgFreeIds :: SpecArg -> IdSet
specArgFreeIds :: SpecArg -> VarSet
specArgFreeIds (SpecType {}) = VarSet
emptyVarSet
specArgFreeIds (SpecDict OutExpr
dx) = OutExpr -> VarSet
exprFreeIds OutExpr
dx
specArgFreeIds SpecArg
UnspecType    = VarSet
emptyVarSet
specArgFreeIds SpecArg
UnspecArg     = VarSet
emptyVarSet

specArgFreeVars :: SpecArg -> VarSet
specArgFreeVars :: SpecArg -> VarSet
specArgFreeVars (SpecType Kind
ty) = Kind -> VarSet
tyCoVarsOfType Kind
ty
specArgFreeVars (SpecDict OutExpr
dx) = OutExpr -> VarSet
exprFreeVars OutExpr
dx
specArgFreeVars SpecArg
UnspecType    = VarSet
emptyVarSet
specArgFreeVars SpecArg
UnspecArg     = VarSet
emptyVarSet

isSpecDict :: SpecArg -> Bool
isSpecDict :: SpecArg -> Bool
isSpecDict (SpecDict {}) = Bool
True
isSpecDict SpecArg
_             = Bool
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 dShowB 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, dShowB :-> 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, dShowB :-> 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 :: SpecEnv
-> [Id]
-> [SpecArg]
-> SpecM
     (Bool, SpecEnv, [Id], [Id], [OutExpr], [Id], [DictBind], [OutExpr])
specHeader SpecEnv
env (Id
bndr : [Id]
bndrs) (SpecType Kind
ty : [SpecArg]
args)
  = do { -- Find qvars, the type variables to add to the binders for the rule
         -- Namely those free in `ty` that aren't in scope
         -- See (MP2) in Note [Specialising polymorphic dictionaries]
         let in_scope :: InScopeSet
in_scope = Subst -> InScopeSet
Core.substInScopeSet (SpecEnv -> Subst
se_subst SpecEnv
env)
             qvars :: [Id]
qvars    = [Id] -> [Id]
scopedSort ([Id] -> [Id]) -> [Id] -> [Id]
forall a b. (a -> b) -> a -> b
$
                        (Id -> Bool) -> [Id] -> [Id]
forall a. (a -> Bool) -> [a] -> [a]
filterOut (Id -> InScopeSet -> Bool
`elemInScopeSet` InScopeSet
in_scope) ([Id] -> [Id]) -> [Id] -> [Id]
forall a b. (a -> b) -> a -> b
$
                        Kind -> [Id]
tyCoVarsOfTypeList Kind
ty
             (SpecEnv
env1, [Id]
qvars') = SpecEnv -> [Id] -> (SpecEnv, [Id])
forall (f :: * -> *).
Traversable f =>
SpecEnv -> f Id -> (SpecEnv, f Id)
substBndrs SpecEnv
env [Id]
qvars
             ty' :: Kind
ty'            = SpecEnv -> Kind -> Kind
substTy SpecEnv
env1 Kind
ty
             env2 :: SpecEnv
env2           = SpecEnv -> Id -> Kind -> SpecEnv
extendTvSubst SpecEnv
env1 Id
bndr Kind
ty'
       ; (useful, env3, leftover_bndrs, rule_bs, rule_es, bs', dx, spec_args)
            <- SpecEnv
-> [Id]
-> [SpecArg]
-> SpecM
     (Bool, SpecEnv, [Id], [Id], [OutExpr], [Id], [DictBind], [OutExpr])
specHeader SpecEnv
env2 [Id]
bndrs [SpecArg]
args
       ; pure ( useful
              , env3
              , leftover_bndrs
              , qvars' ++ rule_bs
              , Type ty' : rule_es
              , qvars' ++ bs'
              , dx
              , Type ty' : 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 SpecEnv
env (Id
bndr : [Id]
bndrs) (SpecArg
UnspecType : [SpecArg]
args)
  = do { let (SpecEnv
env', Id
bndr') = SpecEnv -> Id -> (SpecEnv, Id)
substBndr SpecEnv
env Id
bndr
       ; (useful, env'', leftover_bndrs, rule_bs, rule_es, bs', dx, spec_args)
            <- SpecEnv
-> [Id]
-> [SpecArg]
-> SpecM
     (Bool, SpecEnv, [Id], [Id], [OutExpr], [Id], [DictBind], [OutExpr])
specHeader SpecEnv
env' [Id]
bndrs [SpecArg]
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 SpecEnv
env (Id
bndr : [Id]
bndrs) (SpecDict OutExpr
d : [SpecArg]
args)
  | Bool -> Bool
not (Id -> Bool
isDeadBinder Id
bndr)
  , (Id -> Bool) -> VarSet -> Bool
allVarSet (Id -> InScopeSet -> Bool
`elemInScopeSet` InScopeSet
in_scope) (OutExpr -> VarSet
exprFreeVars OutExpr
d)
    -- See Note [Weird special case for SpecDict]
  = do { (env1, bndr') <- SpecEnv -> Id -> UniqSM (SpecEnv, Id)
newDictBndr SpecEnv
env Id
bndr -- See Note [Zap occ info in rule binders]
       ; let (env2, dx_bind, spec_dict) = bindAuxiliaryDict env1 bndr bndr' d
       ; (_, env3, leftover_bndrs, rule_bs, rule_es, bs', dx, spec_args)
             <- specHeader env2 bndrs args
       ; pure ( True      -- Ha!  A useful specialisation!
              , env3
              , leftover_bndrs
              -- See Note [Evidence foralls]
              , exprFreeIdsList (varToCoreExpr bndr') ++ rule_bs
              , varToCoreExpr bndr' : rule_es
              , bs'
              , maybeToList dx_bind ++ dx
              , spec_dict : spec_args
              )
       }
   where
     in_scope :: InScopeSet
in_scope = Subst -> InScopeSet
Core.substInScopeSet (SpecEnv -> Subst
se_subst SpecEnv
env)

-- Finally, we don't want to specialise on this argument 'i':
--   - It's an UnSpecArg, or
--   - It's a dead dictionary
-- 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 SpecEnv
env (Id
bndr : [Id]
bndrs) (SpecArg
_ : [SpecArg]
args)
    -- The "_" can be UnSpecArg, or SpecDict where the bndr is dead
  = do { -- see Note [Zap occ info in rule binders]
         let (SpecEnv
env', Id
bndr') = SpecEnv -> Id -> (SpecEnv, Id)
substBndr SpecEnv
env (Id -> Id
zapIdOccInfo Id
bndr)
       ; (useful, env'', leftover_bndrs, rule_bs, rule_es, bs', dx, spec_args)
             <- SpecEnv
-> [Id]
-> [SpecArg]
-> SpecM
     (Bool, SpecEnv, [Id], [Id], [OutExpr], [Id], [DictBind], [OutExpr])
specHeader SpecEnv
env' [Id]
bndrs [SpecArg]
args

       ; let bndr_ty = Id -> Kind
idType Id
bndr'

             -- See Note [Drop dead args from specialisations]
             -- C.f. GHC.Core.Opt.WorkWrap.Utils.mk_absent_let
             (mb_spec_bndr, spec_arg)
                | isDeadBinder bndr
                , Just lit_expr <- mkLitRubbish bndr_ty
                = (Nothing, lit_expr)
                | otherwise
                = (Just bndr', varToCoreExpr bndr')

       ; pure ( useful
              , env''
              , leftover_bndrs
              , bndr' : rule_bs
              , varToCoreExpr bndr' : rule_es
              , case mb_spec_bndr of
                  Just Id
b' -> Id
b' Id -> [Id] -> [Id]
forall a. a -> [a] -> [a]
: [Id]
bs'
                  Maybe Id
Nothing -> [Id]
bs'
              , dx
              , spec_arg : spec_args
              )
       }

-- If we run out of binders, stop immediately
-- See Note [Specialisation Must Preserve Sharing]
specHeader SpecEnv
env [] [SpecArg]
_ = (Bool, SpecEnv, [Id], [Id], [OutExpr], [Id], [DictBind], [OutExpr])
-> SpecM
     (Bool, SpecEnv, [Id], [Id], [OutExpr], [Id], [DictBind], [OutExpr])
forall a. a -> UniqSM a
forall (f :: * -> *) a. Applicative f => a -> f a
pure (Bool
False, SpecEnv
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 SpecEnv
env [Id]
bndrs []
  = (Bool, SpecEnv, [Id], [Id], [OutExpr], [Id], [DictBind], [OutExpr])
-> SpecM
     (Bool, SpecEnv, [Id], [Id], [OutExpr], [Id], [DictBind], [OutExpr])
forall a. a -> UniqSM a
forall (f :: * -> *) a. Applicative f => a -> f a
pure (Bool
False, SpecEnv
env', [Id]
bndrs', [], [], [], [], [])
  where
    (SpecEnv
env', [Id]
bndrs') = SpecEnv -> [Id] -> (SpecEnv, [Id])
forall (f :: * -> *).
Traversable f =>
SpecEnv -> f Id -> (SpecEnv, f Id)
substBndrs SpecEnv
env [Id]
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        -- Substitutes for orig_dict_id
     , Maybe DictBind -- Auxiliary dict binding, if any
     , OutExpr)       -- Witnessing expression (always trivial)
bindAuxiliaryDict :: SpecEnv
-> Id -> Id -> OutExpr -> (SpecEnv, Maybe DictBind, OutExpr)
bindAuxiliaryDict env :: SpecEnv
env@(SE { se_subst :: SpecEnv -> Subst
se_subst = Subst
subst })
                  Id
orig_dict_id Id
fresh_dict_id OutExpr
dict_expr

  -- If the dictionary argument is trivial,
  -- don’t bother creating a new dict binding; just substitute
  | OutExpr -> Bool
exprIsTrivial OutExpr
dict_expr
  = let env' :: SpecEnv
env' = SpecEnv
env { se_subst = Core.extendSubst subst orig_dict_id dict_expr }
    in -- pprTrace "bindAuxiliaryDict:trivial" (ppr orig_dict_id <+> ppr dict_id) $
       (SpecEnv
env', Maybe DictBind
forall a. Maybe a
Nothing, OutExpr
dict_expr)

  | Bool
otherwise  -- Non-trivial dictionary arg; make an auxiliary binding
  = let fresh_dict_id' :: Id
fresh_dict_id' = Id
fresh_dict_id Id -> OutExpr -> Id
`addDictUnfolding` OutExpr
dict_expr

        dict_bind :: DictBind
dict_bind = InBind -> DictBind
mkDB (Id -> OutExpr -> InBind
forall b. b -> Expr b -> Bind b
NonRec Id
fresh_dict_id' OutExpr
dict_expr)
        env' :: SpecEnv
env' = SpecEnv
env { se_subst = Core.extendSubst subst orig_dict_id (Var fresh_dict_id')
                                `Core.extendSubstInScope` fresh_dict_id' }
                                -- Ensure the new unfolding is in the in-scope set
    in -- pprTrace "bindAuxiliaryDict:non-trivial" (ppr orig_dict_id <+> ppr fresh_dict_id') $
       (SpecEnv
env', DictBind -> Maybe DictBind
forall a. a -> Maybe a
Just DictBind
dict_bind, Id -> OutExpr
forall b. Id -> Expr b
Var Id
fresh_dict_id')

addDictUnfolding :: Id -> CoreExpr -> Id
-- Add unfolding for freshly-bound Ids: see Note [Make the new dictionaries interesting]
-- and Note [Specialisation modulo dictionary selectors]
addDictUnfolding :: Id -> OutExpr -> Id
addDictUnfolding Id
id OutExpr
rhs
  = Id
id Id -> Unfolding -> Id
`setIdUnfolding` UnfoldingOpts -> OutExpr -> Unfolding
mkSimpleUnfolding UnfoldingOpts
defaultUnfoldingOpts OutExpr
rhs

{-
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; so we add an unfolding
in the dictionary Id.
-}


{- *********************************************************************
*                                                                      *
            UsageDetails and suchlike
*                                                                      *
********************************************************************* -}

data UsageDetails
  = MkUD { UsageDetails -> FloatedDictBinds
ud_binds :: !FloatedDictBinds
         , UsageDetails -> CallDetails
ud_calls :: !CallDetails }
    -- INVARIANT: suppose bs = fdb_bndrs ud_binds
    -- Then 'calls' may *mention* 'bs',
    -- but there should be no calls *for* bs

data FloatedDictBinds  -- See Note [Floated dictionary bindings]
  = FDB { FloatedDictBinds -> OrdList DictBind
fdb_binds :: !(OrdList DictBind)
               -- The order is important;
               -- in ds1 `appOL` ds2, bindings in ds2 can depend on those in ds1

        , FloatedDictBinds -> VarSet
fdb_bndrs :: !IdSet
    }          -- ^ The binders of 'fdb_binds'.
               -- Caches a superset of the expression
               --   `mkVarSet (bindersOfDictBinds fdb_binds))`
               -- for later addition to an InScopeSet

-- | A 'DictBind' is a binding along with a cached set containing its free
-- variables (both type variables and dictionaries). We need this set
-- in splitDictBinds, when filtering bindings to decide which are
-- captured by a binder
data DictBind = DB { DictBind -> InBind
db_bind :: CoreBind, DictBind -> VarSet
db_fvs :: VarSet }

bindersOfDictBind :: DictBind -> [Id]
bindersOfDictBind :: DictBind -> [Id]
bindersOfDictBind = InBind -> [Id]
forall b. Bind b -> [b]
bindersOf (InBind -> [Id]) -> (DictBind -> InBind) -> DictBind -> [Id]
forall b c a. (b -> c) -> (a -> b) -> a -> c
. DictBind -> InBind
db_bind

bindersOfDictBinds :: Foldable f => f DictBind -> [Id]
bindersOfDictBinds :: forall (f :: * -> *). Foldable f => f DictBind -> [Id]
bindersOfDictBinds = CoreProgram -> [Id]
forall b. [Bind b] -> [b]
bindersOfBinds (CoreProgram -> [Id])
-> (f DictBind -> CoreProgram) -> f DictBind -> [Id]
forall b c a. (b -> c) -> (a -> b) -> a -> c
. (DictBind -> CoreProgram -> CoreProgram)
-> CoreProgram -> f DictBind -> CoreProgram
forall a b. (a -> b -> b) -> b -> f a -> b
forall (t :: * -> *) a b.
Foldable t =>
(a -> b -> b) -> b -> t a -> b
foldr ((:) (InBind -> CoreProgram -> CoreProgram)
-> (DictBind -> InBind) -> DictBind -> CoreProgram -> CoreProgram
forall b c a. (b -> c) -> (a -> b) -> a -> c
. DictBind -> InBind
db_bind) []

{- 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 :: OrdList DictBind) may contain
non-dictionary bindings too.

Note [Specialising polymorphic dictionaries]
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Note June 2023: This has proved to be quite a tricky optimisation to get right
see (#23469, #23109, #21229, #23445) so it is now guarded by a flag
`-fpolymorphic-specialisation`.

Consider
    class M a where { foo :: a -> Int }

    instance M (ST s) where ...
    -- dMST :: forall s. M (ST s)

    wimwam :: forall a. M a => a -> Int
    wimwam = /\a \(d::M a). body

    f :: ST s -> Int
    f = /\s \(x::ST s). wimwam @(ST s) (dMST @s) dx + 1

We'd like to specialise wimwam at (ST s), thus
    $swimwam :: forall s. ST s -> Int
    $swimwam = /\s. body[ST s/a, (dMST @s)/d]

    RULE forall s (d :: M (ST s)).
         wimwam @(ST s) d = $swimwam @s

Here are the moving parts:

(MP1) We must /not/ dump the CallInfo
        CIS wimwam (CI { ci_key = [@(ST s), dMST @s]
                       , ci_fvs = {dMST} })
      when we come to the /\s.  Instead, we simply let it continue to float
      upwards. Hence ci_fvs is an IdSet, listing the /Ids/ that
      are free in the call, but not the /TyVars/.  Hence using specArgFreeIds
      in singleCall.

  NB to be fully kosher we should explicitly quantifying the CallInfo
  over 's', but we don't bother.  This would matter if there was an
  enclosing binding of the same 's', which I don't expect to happen.

(MP2) When we come to specialise the call, we must remember to quantify
      over 's'.  That is done in the SpecType case of specHeader, where
      we add 's' (called qvars) to the binders of the RULE and the specialised
      function.

(MP3) If we have f :: forall m. Monoid m => blah, and two calls
        (f @(Endo b)      (d1 :: Monoid (Endo b))
        (f @(Endo (c->c)) (d2 :: Monoid (Endo (c->c)))
      we want to generate a specialisation only for the first.  The second
      is just a substitution instance of the first, with no greater specialisation.
      Hence the use of `removeDupCalls` in `filterCalls`.

      You might wonder if `d2` might be more specialised than `d1`; but no.
      This `removeDupCalls` thing is at the definition site of `f`, and both `d1`
      and `d2` are in scope. So `d1` is simply more polymorphic than `d2`, but
      is just as specialised.

      This distinction is sadly lost once we build a RULE, so `alreadyCovered`
      can't be so clever.  E.g if we have an existing RULE
            forall @a (d1:Ord Int) (d2: Eq a). f @a @Int d1 d2 = ...
      and a putative new rule
            forall (d1:Ord Int) (d2: Eq Int). f @Int @Int d1 d2 = ...
      we /don't/ want the existing rule to subsume the new one.

      So we sadly put up with having two rather different places where we
      eliminate duplicates: `alreadyCovered` and `removeDupCalls`.

All this arose in #13873, in the unexpected form that a SPECIALISE
pragma made the program slower!  The reason was that the specialised
function $sinsertWith arising from the pragma looked rather like `f`
above, and failed to specialise a call in its body like wimwam.
Without the pragma, the original call to `insertWith` was completely
monomorpic, and specialised in one go.

Wrinkles.

* See Note [Weird special case for SpecDict]

* With -XOverlappingInstances you might worry about this:
    class C a where ...
    instance C (Maybe Int) where ...   -- $df1 :: C (Maybe Int)
    instance C (Maybe a)   where ...   -- $df2 :: forall a. C (Maybe a)

    f :: C a => blah
    f = rhs

    g = /\a.  ...(f @(Maybe a) ($df2 a))...
    h = ...f @(Maybe Int) $df1

  There are two calls to f, but with different evidence.  This patch will
  combine them into one.  But it's OK: this code will never arise unless you
  use -XIncoherentInstances.  Even with -XOverlappingInstances, GHC tries hard
  to keep dictionaries as singleton types.  But that goes out of the window
  with -XIncoherentInstances -- and that is true even with ordianry type-class
  specialisation (at least if any inlining has taken place).

  GHC makes very few guarantees when you use -XIncoherentInstances, and its
  not worth crippling the normal case for the incoherent corner.  (The best
  thing might be to switch off specialisation altogether if incoherence is
  involved... but incoherence is a property of an instance, not a class, so
  it's a hard test to make.)

  But see Note [Specialisation and overlapping instances].

Note [Weird special case for SpecDict]
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Suppose we are trying to specialise for this this call:
   $wsplit @T (mkD @k @(a::k) :: C T)
where
   mkD :: forall k (a::k). C T
is a top-level dictionary-former.  This actually happened in #22459,
because of (MP1) of Note [Specialising polymorphic dictionaries].

How can we specialise $wsplit?  We might try

   RULE "SPEC" forall (d :: C T). $wsplit @T d = $s$wsplit

but then in the body of $s$wsplit what will we use for the dictionary
evidence?  We can't use (mkD @k @(a::k)) because k and a aren't in scope.
We could zap `k` to (Any @Type) and `a` to (Any @(Any @Type)), but that
is a lot of hard work for a very strange case.

So we simply refrain from specialising in this case; hence the guard
   allVarSet (`elemInScopeSet` in_scope) (exprFreeVars d)
in the SpecDict cased of specHeader.

How did this strange polymorphic mkD arise in the first place?
From GHC.Core.Opt.Utils.abstractFloats, which was abstracting
over too many type variables. But that too is now fixed;
see Note [Which type variables to abstract over] in that module.
-}

instance Outputable DictBind where
  ppr :: DictBind -> SDoc
ppr (DB { db_bind :: DictBind -> InBind
db_bind = InBind
bind, db_fvs :: DictBind -> VarSet
db_fvs = VarSet
fvs })
    = String -> SDoc
forall doc. IsLine doc => String -> doc
text String
"DB" SDoc -> SDoc -> SDoc
forall doc. IsLine doc => doc -> doc -> doc
<+> SDoc -> SDoc
forall doc. IsLine doc => doc -> doc
braces ([SDoc] -> SDoc
forall doc. IsLine doc => [doc] -> doc
sep [ String -> SDoc
forall doc. IsLine doc => String -> doc
text String
"fvs: " SDoc -> SDoc -> SDoc
forall doc. IsLine doc => doc -> doc -> doc
<+> VarSet -> SDoc
forall a. Outputable a => a -> SDoc
ppr VarSet
fvs
                                , String -> SDoc
forall doc. IsLine doc => String -> doc
text String
"bind:" SDoc -> SDoc -> SDoc
forall doc. IsLine doc => doc -> doc -> doc
<+> InBind -> SDoc
forall a. Outputable a => a -> SDoc
ppr InBind
bind ])

instance Outputable UsageDetails where
  ppr :: UsageDetails -> SDoc
ppr (MkUD { ud_binds :: UsageDetails -> FloatedDictBinds
ud_binds = FloatedDictBinds
dbs, ud_calls :: UsageDetails -> CallDetails
ud_calls = CallDetails
calls })
        = String -> SDoc
forall doc. IsLine doc => String -> doc
text String
"MkUD" SDoc -> SDoc -> SDoc
forall doc. IsLine doc => doc -> doc -> doc
<+> SDoc -> SDoc
forall doc. IsLine doc => doc -> doc
braces ([SDoc] -> SDoc
forall doc. IsLine doc => [doc] -> doc
sep (SDoc -> [SDoc] -> [SDoc]
forall doc. IsLine doc => doc -> [doc] -> [doc]
punctuate SDoc
forall doc. IsLine doc => doc
comma
                [String -> SDoc
forall doc. IsLine doc => String -> doc
text String
"binds" SDoc -> SDoc -> SDoc
forall doc. IsLine doc => doc -> doc -> doc
<+> SDoc
forall doc. IsLine doc => doc
equals SDoc -> SDoc -> SDoc
forall doc. IsLine doc => doc -> doc -> doc
<+> FloatedDictBinds -> SDoc
forall a. Outputable a => a -> SDoc
ppr FloatedDictBinds
dbs,
                 String -> SDoc
forall doc. IsLine doc => String -> doc
text String
"calls" SDoc -> SDoc -> SDoc
forall doc. IsLine doc => doc -> doc -> doc
<+> SDoc
forall doc. IsLine doc => doc
equals SDoc -> SDoc -> SDoc
forall doc. IsLine doc => doc -> doc -> doc
<+> CallDetails -> SDoc
forall a. Outputable a => a -> SDoc
ppr CallDetails
calls]))

instance Outputable FloatedDictBinds where
  ppr :: FloatedDictBinds -> SDoc
ppr (FDB { fdb_binds :: FloatedDictBinds -> OrdList DictBind
fdb_binds = OrdList DictBind
binds }) = OrdList DictBind -> SDoc
forall a. Outputable a => a -> SDoc
ppr OrdList DictBind
binds

emptyUDs :: UsageDetails
emptyUDs :: UsageDetails
emptyUDs = MkUD { ud_binds :: FloatedDictBinds
ud_binds = FloatedDictBinds
emptyFDBs, ud_calls :: CallDetails
ud_calls = CallDetails
forall a. DVarEnv a
emptyDVarEnv }


emptyFDBs :: FloatedDictBinds
emptyFDBs :: FloatedDictBinds
emptyFDBs = FDB { fdb_binds :: OrdList DictBind
fdb_binds = OrdList DictBind
forall a. OrdList a
nilOL, fdb_bndrs :: VarSet
fdb_bndrs = VarSet
emptyVarSet }

------------------------------------------------------------
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 alreadyCovered in specCalls

data CallInfo
  = CI { CallInfo -> [SpecArg]
ci_key  :: [SpecArg]   -- Arguments of the call
                                -- See Note [The (CI-KEY) invariant]

       , CallInfo -> VarSet
ci_fvs  :: IdSet       -- Free Ids of the ci_key call
                                -- /not/ including the main id itself, of course
                                -- NB: excluding tyvars:
                                --     See Note [Specialising polymorphic dictionaries]
    }

{- Note [The (CI-KEY) invariant]
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Invariant (CI-KEY):
   In the `ci_key :: [SpecArg]` field of `CallInfo`,
     * The list is non-empty
     * The least element is always a `SpecDict`

In this way the RULE has as few args as possible, which broadens its
applicability, since rules only fire when saturated.
-}

type DictExpr = CoreExpr

ciSetFilter :: (CallInfo -> Bool) -> CallInfoSet -> CallInfoSet
ciSetFilter :: (CallInfo -> Bool) -> CallInfoSet -> CallInfoSet
ciSetFilter CallInfo -> Bool
p (CIS Id
id Bag CallInfo
a) = Id -> Bag CallInfo -> CallInfoSet
CIS Id
id ((CallInfo -> Bool) -> Bag CallInfo -> Bag CallInfo
forall a. (a -> Bool) -> Bag a -> Bag a
filterBag CallInfo -> Bool
p Bag CallInfo
a)

instance Outputable CallInfoSet where
  ppr :: CallInfoSet -> SDoc
ppr (CIS Id
fn Bag CallInfo
map) = SDoc -> Int -> SDoc -> SDoc
hang (String -> SDoc
forall doc. IsLine doc => String -> doc
text String
"CIS" SDoc -> SDoc -> SDoc
forall doc. IsLine doc => doc -> doc -> doc
<+> Id -> SDoc
forall a. Outputable a => a -> SDoc
ppr Id
fn)
                        Int
2 (Bag CallInfo -> SDoc
forall a. Outputable a => a -> SDoc
ppr Bag CallInfo
map)

pprCallInfo :: Id -> CallInfo -> SDoc
pprCallInfo :: Id -> CallInfo -> SDoc
pprCallInfo Id
fn (CI { ci_key :: CallInfo -> [SpecArg]
ci_key = [SpecArg]
key })
  = Id -> SDoc
forall a. Outputable a => a -> SDoc
ppr Id
fn SDoc -> SDoc -> SDoc
forall doc. IsLine doc => doc -> doc -> doc
<+> [SpecArg] -> SDoc
forall a. Outputable a => a -> SDoc
ppr [SpecArg]
key

instance Outputable CallInfo where
  ppr :: CallInfo -> SDoc
ppr (CI { ci_key :: CallInfo -> [SpecArg]
ci_key = [SpecArg]
key, ci_fvs :: CallInfo -> VarSet
ci_fvs = VarSet
_fvs })
    = String -> SDoc
forall doc. IsLine doc => String -> doc
text String
"CI" SDoc -> SDoc -> SDoc
forall doc. IsLine doc => doc -> doc -> doc
<> SDoc -> SDoc
forall doc. IsLine doc => doc -> doc
braces ([SDoc] -> SDoc
forall doc. IsLine doc => [doc] -> doc
sep ((SpecArg -> SDoc) -> [SpecArg] -> [SDoc]
forall a b. (a -> b) -> [a] -> [b]
map SpecArg -> SDoc
forall a. Outputable a => a -> SDoc
ppr [SpecArg]
key))

unionCalls :: CallDetails -> CallDetails -> CallDetails
unionCalls :: CallDetails -> CallDetails -> CallDetails
unionCalls CallDetails
c1 CallDetails
c2 = (CallInfoSet -> CallInfoSet -> CallInfoSet)
-> CallDetails -> CallDetails -> CallDetails
forall a. (a -> a -> a) -> DVarEnv a -> DVarEnv a -> DVarEnv a
plusDVarEnv_C CallInfoSet -> CallInfoSet -> CallInfoSet
unionCallInfoSet CallDetails
c1 CallDetails
c2

unionCallInfoSet :: CallInfoSet -> CallInfoSet -> CallInfoSet
unionCallInfoSet :: CallInfoSet -> CallInfoSet -> CallInfoSet
unionCallInfoSet (CIS Id
f Bag CallInfo
calls1) (CIS Id
_ Bag CallInfo
calls2) =
  Id -> Bag CallInfo -> CallInfoSet
CIS Id
f (Bag CallInfo
calls1 Bag CallInfo -> Bag CallInfo -> Bag CallInfo
forall a. Bag a -> Bag a -> Bag a
`unionBags` Bag CallInfo
calls2)

callDetailsFVs :: CallDetails -> VarSet
callDetailsFVs :: CallDetails -> VarSet
callDetailsFVs CallDetails
calls =
  (CallInfoSet -> VarSet -> VarSet)
-> VarSet -> CallDetails -> VarSet
forall {k} elt a (key :: k).
(elt -> a -> a) -> a -> UniqDFM key elt -> a
nonDetStrictFoldUDFM (VarSet -> VarSet -> VarSet
unionVarSet (VarSet -> VarSet -> VarSet)
-> (CallInfoSet -> VarSet) -> CallInfoSet -> VarSet -> VarSet
forall b c a. (b -> c) -> (a -> b) -> a -> c
. CallInfoSet -> VarSet
callInfoFVs) VarSet
emptyVarSet CallDetails
calls
  -- It's OK to use nonDetStrictFoldUDFM here because we forget the ordering
  -- immediately by converting to a nondeterministic set.

callInfoFVs :: CallInfoSet -> VarSet
callInfoFVs :: CallInfoSet -> VarSet
callInfoFVs (CIS Id
_ Bag CallInfo
call_info) =
  (CallInfo -> VarSet -> VarSet) -> VarSet -> Bag CallInfo -> VarSet
forall a b. (a -> b -> b) -> b -> Bag a -> b
forall (t :: * -> *) a b.
Foldable t =>
(a -> b -> b) -> b -> t a -> b
foldr (\(CI { ci_fvs :: CallInfo -> VarSet
ci_fvs = VarSet
fv }) VarSet
vs -> VarSet -> VarSet -> VarSet
unionVarSet VarSet
fv VarSet
vs) VarSet
emptyVarSet Bag CallInfo
call_info

getTheta :: [PiTyBinder] -> [PredType]
getTheta :: [PiTyBinder] -> [Kind]
getTheta = (PiTyBinder -> Kind) -> [PiTyBinder] -> [Kind]
forall a b. (a -> b) -> [a] -> [b]
forall (f :: * -> *) a b. Functor f => (a -> b) -> f a -> f b
fmap PiTyBinder -> Kind
piTyBinderType ([PiTyBinder] -> [Kind])
-> ([PiTyBinder] -> [PiTyBinder]) -> [PiTyBinder] -> [Kind]
forall b c a. (b -> c) -> (a -> b) -> a -> c
. (PiTyBinder -> Bool) -> [PiTyBinder] -> [PiTyBinder]
forall a. (a -> Bool) -> [a] -> [a]
filter PiTyBinder -> Bool
isInvisiblePiTyBinder ([PiTyBinder] -> [PiTyBinder])
-> ([PiTyBinder] -> [PiTyBinder]) -> [PiTyBinder] -> [PiTyBinder]
forall b c a. (b -> c) -> (a -> b) -> a -> c
. (PiTyBinder -> Bool) -> [PiTyBinder] -> [PiTyBinder]
forall a. (a -> Bool) -> [a] -> [a]
filter PiTyBinder -> Bool
isAnonPiTyBinder


------------------------------------------------------------
singleCall :: SpecEnv -> Id -> [SpecArg] -> UsageDetails
singleCall :: SpecEnv -> Id -> [SpecArg] -> UsageDetails
singleCall SpecEnv
spec_env Id
id [SpecArg]
args
  = MkUD {ud_binds :: FloatedDictBinds
ud_binds = FloatedDictBinds
emptyFDBs,
          ud_calls :: CallDetails
ud_calls = Id -> CallInfoSet -> CallDetails
forall a. Id -> a -> DVarEnv a
unitDVarEnv Id
id (CallInfoSet -> CallDetails) -> CallInfoSet -> CallDetails
forall a b. (a -> b) -> a -> b
$ Id -> Bag CallInfo -> CallInfoSet
CIS Id
id (Bag CallInfo -> CallInfoSet) -> Bag CallInfo -> CallInfoSet
forall a b. (a -> b) -> a -> b
$
                     CallInfo -> Bag CallInfo
forall a. a -> Bag a
unitBag (CI { ci_key :: [SpecArg]
ci_key  = [SpecArg]
args
                                 , ci_fvs :: VarSet
ci_fvs  = VarSet
call_fvs }) }
  where
    call_fvs :: VarSet
call_fvs =
      (SpecArg -> VarSet -> VarSet) -> VarSet -> [SpecArg] -> VarSet
forall a b. (a -> b -> b) -> b -> [a] -> b
forall (t :: * -> *) a b.
Foldable t =>
(a -> b -> b) -> b -> t a -> b
foldr (VarSet -> VarSet -> VarSet
unionVarSet (VarSet -> VarSet -> VarSet)
-> (SpecArg -> VarSet) -> SpecArg -> VarSet -> VarSet
forall b c a. (b -> c) -> (a -> b) -> a -> c
. SpecArg -> VarSet
free_var_fn) VarSet
emptyVarSet [SpecArg]
args

    free_var_fn :: SpecArg -> VarSet
free_var_fn =
      if GeneralFlag -> DynFlags -> Bool
gopt GeneralFlag
Opt_PolymorphicSpecialisation (SpecEnv -> DynFlags
se_dflags SpecEnv
spec_env)
        then SpecArg -> VarSet
specArgFreeIds
        else SpecArg -> VarSet
specArgFreeVars



        -- specArgFreeIds: we specifically look for free Ids, not TyVars
        --    see (MP1) in Note [Specialising polymorphic dictionaries]
        --
        -- We don't include the 'id' itself.

mkCallUDs :: SpecEnv -> OutExpr -> [OutExpr] -> UsageDetails
mkCallUDs :: SpecEnv -> OutExpr -> [OutExpr] -> UsageDetails
mkCallUDs SpecEnv
env OutExpr
fun [OutExpr]
args
  | ([CoreTickish]
_, Var Id
f) <- (CoreTickish -> Bool) -> OutExpr -> ([CoreTickish], OutExpr)
forall b.
(CoreTickish -> Bool) -> Expr b -> ([CoreTickish], Expr b)
stripTicksTop CoreTickish -> Bool
forall (pass :: TickishPass). GenTickish pass -> Bool
tickishFloatable OutExpr
fun -- See Note [Ticks on applications]
  = -- pprTraceWith "mkCallUDs" (\res -> vcat [ ppr f, ppr args, ppr res ]) $
    SpecEnv -> Id -> [OutExpr] -> UsageDetails
mkCallUDs' SpecEnv
env Id
f [OutExpr]
args
  | Bool
otherwise
  = UsageDetails
emptyUDs

mkCallUDs' :: SpecEnv -> Id -> [OutExpr] -> UsageDetails
mkCallUDs' :: SpecEnv -> Id -> [OutExpr] -> UsageDetails
mkCallUDs' SpecEnv
env Id
f [OutExpr]
args
  | SpecEnv -> Id -> Bool
wantCallsFor SpecEnv
env Id
f    -- We want it, and...
  , Bool -> Bool
not ([SpecArg] -> Bool
forall a. [a] -> Bool
forall (t :: * -> *) a. Foldable t => t a -> Bool
null [SpecArg]
ci_key)     -- this call site has a useful specialisation
  = -- pprTrace "mkCallUDs: keeping" _trace_doc
    SpecEnv -> Id -> [SpecArg] -> UsageDetails
singleCall SpecEnv
env Id
f [SpecArg]
ci_key

  | Bool
otherwise  -- See also Note [Specialisations already covered]
  = -- pprTrace "mkCallUDs: discarding" _trace_doc
    UsageDetails
emptyUDs

  where
    _trace_doc :: SDoc
_trace_doc = [SDoc] -> SDoc
forall doc. IsDoc doc => [doc] -> doc
vcat [Id -> SDoc
forall a. Outputable a => a -> SDoc
ppr Id
f, [OutExpr] -> SDoc
forall a. Outputable a => a -> SDoc
ppr [OutExpr]
args, [SpecArg] -> SDoc
forall a. Outputable a => a -> SDoc
ppr [SpecArg]
ci_key]
    pis :: [PiTyBinder]
pis                = ([PiTyBinder], Kind) -> [PiTyBinder]
forall a b. (a, b) -> a
fst (([PiTyBinder], Kind) -> [PiTyBinder])
-> ([PiTyBinder], Kind) -> [PiTyBinder]
forall a b. (a -> b) -> a -> b
$ Kind -> ([PiTyBinder], Kind)
splitPiTys (Kind -> ([PiTyBinder], Kind)) -> Kind -> ([PiTyBinder], Kind)
forall a b. (a -> b) -> a -> b
$ Id -> Kind
idType Id
f
    constrained_tyvars :: VarSet
constrained_tyvars = [Kind] -> VarSet
tyCoVarsOfTypes ([Kind] -> VarSet) -> [Kind] -> VarSet
forall a b. (a -> b) -> a -> b
$ [PiTyBinder] -> [Kind]
getTheta [PiTyBinder]
pis

    ci_key :: [SpecArg]
    ci_key :: [SpecArg]
ci_key = (SpecArg -> Bool) -> [SpecArg] -> [SpecArg]
forall a. (a -> Bool) -> [a] -> [a]
dropWhileEndLE (Bool -> Bool
not (Bool -> Bool) -> (SpecArg -> Bool) -> SpecArg -> Bool
forall b c a. (b -> c) -> (a -> b) -> a -> c
. SpecArg -> Bool
isSpecDict) ([SpecArg] -> [SpecArg]) -> [SpecArg] -> [SpecArg]
forall a b. (a -> b) -> a -> b
$
             (OutExpr -> PiTyBinder -> SpecArg)
-> [OutExpr] -> [PiTyBinder] -> [SpecArg]
forall a b c. (a -> b -> c) -> [a] -> [b] -> [c]
zipWith OutExpr -> PiTyBinder -> SpecArg
mk_spec_arg [OutExpr]
args [PiTyBinder]
pis
             -- Establish (CI-KEY): drop trailing args until we get to a SpecDict

    mk_spec_arg :: OutExpr -> PiTyBinder -> SpecArg
    mk_spec_arg :: OutExpr -> PiTyBinder -> SpecArg
mk_spec_arg OutExpr
arg (Named ForAllTyBinder
bndr)
      |  ForAllTyBinder -> Id
forall tv argf. VarBndr tv argf -> tv
binderVar ForAllTyBinder
bndr Id -> VarSet -> Bool
`elemVarSet` VarSet
constrained_tyvars
      = case OutExpr
arg of
          Type Kind
ty -> Kind -> SpecArg
SpecType Kind
ty
          OutExpr
_       -> String -> SDoc -> SpecArg
forall a. HasCallStack => String -> SDoc -> a
pprPanic String
"ci_key" (SDoc -> SpecArg) -> SDoc -> SpecArg
forall a b. (a -> b) -> a -> b
$ OutExpr -> SDoc
forall a. Outputable a => a -> SDoc
ppr OutExpr
arg
      |  Bool
otherwise = SpecArg
UnspecType

    -- For "invisibleFunArg", 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 OutExpr
arg (Anon Scaled Kind
pred FunTyFlag
af)
      | FunTyFlag -> Bool
isInvisibleFunArg FunTyFlag
af
      , OutExpr -> Kind -> Bool
interestingDict OutExpr
arg (Scaled Kind -> Kind
forall a. Scaled a -> a
scaledThing Scaled Kind
pred)
              -- See Note [Interesting dictionary arguments]
      = OutExpr -> SpecArg
SpecDict OutExpr
arg

      | Bool
otherwise = SpecArg
UnspecArg

{-
Note [Ticks on applications]
~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Ticks such as source location annotations can sometimes make their way
onto applications (see e.g. #21697). So if we see something like

    App (Tick _ f) e

we need to descend below the tick to find what the real function being
applied is.

The resulting RULE also has to be able to match this annotated use
site, so we only look through ticks that RULE matching looks through
(see Note [Tick annotations in RULE matching] in GHC.Core.Rules).
-}

wantCallsFor :: SpecEnv -> Id -> Bool
-- See Note [wantCallsFor]
wantCallsFor :: SpecEnv -> Id -> Bool
wantCallsFor SpecEnv
_env Id
f
  = case Id -> IdDetails
idDetails Id
f of
      RecSelId {}      -> Bool
False
      DataConWorkId {} -> Bool
False
      DataConWrapId {} -> Bool
False
      ClassOpId {}     -> Bool
False
      PrimOpId {}      -> Bool
False
      FCallId {}       -> Bool
False
      TickBoxOpId {}   -> Bool
False
      CoVarId {}       -> Bool
False

      DFunId {}        -> Bool
True
      VanillaId {}     -> Bool
True
      JoinId {}        -> Bool
True
      WorkerLikeId {}  -> Bool
True
      RepPolyId {}     -> Bool
True

{- Note [wantCallsFor]
~~~~~~~~~~~~~~~~~~~~~~
`wantCallsFor env f` says whether the Specialiser should collect calls for
function `f`; other thing being equal, the fewer calls we collect the better. It
is False for things we can't specialise:

* ClassOpId: never inline and we don't have a defn to specialise; we specialise
  them through fireRewriteRules.
* PrimOpId: are never overloaded
* Data constructors: we never specialise them

We could reduce the size of the UsageDetails by being less eager about
collecting calls for some 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 [Interesting dictionary arguments]
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
In `mkCallUDs` we only use `SpecDict` for dictionaries of which
`interestingDict` holds.  Otherwise we use `UnspecArg`.  Two reasons:

* 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.

* Consider this (#25703):
     f :: (Eq a, Show b) => a -> b -> INt
     goo :: forall x. (Eq x) => x -> blah
     goo @x (d:Eq x) (arg:x) = ...(f @x @Int d $fShowInt)...
  If we built a `ci_key` with a (SpecDict d) for `d`, we would end up
  discarding the call at the `\d`.  But if we use `UnspecArg` for that
  uninteresting `d`, we'll get a `ci_key` of
      f @x @Int UnspecArg (SpecDict $fShowInt)
  and /that/ can float out to f's definition and specialise nicely.
  Hooray.  (NB: the call can float only if `-fpolymorphic-specialisation`
  is on; otherwise it'll be trapped by the `\@x -> ...`.)(

What is "interesting"?  (See `interestingDict`.)  Just that it has *some*
structure.  But what about variables?  We look in the variable's /unfolding/.
And that means that we must be careful to ensure that dictionaries /have/
unfoldings,

* cloneBndrSM discards non-Stable unfoldings
* specBind updates the unfolding after specialisation
  See Note [Update unfolding after specialisation]
* bindAuxiliaryDict adds an unfolding for an aux dict
  see Note [Specialisation modulo dictionary selectors]
* specCase adds unfoldings for the new bindings it creates

We accidentally lost accurate tracking of local variables for a long
time, because cloned variables didn'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.

Note [Update unfolding after specialisation]
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Consider (#21848)

  wombat :: Show b => Int -> b -> String
  wombat a b | a>0       = wombat (a-1) b
             | otherwise = show a ++ wombat a b

  class C a where
    meth :: Show b => a -> b -> String
    dummy :: a -> () -- Force a datatype dictionary representation

  instance C Int where
    meth = wombat
    dummy _ = ()

  class C a => D a   -- D has C as a superclass
  instance D Int

  f :: (D a, Show b) => a -> b -> String
  {-# INLINABLE[0] f #-}
  f a b = meth a b ++ "!" ++ meth a b

Now `f` turns into:

  f @a @b (dd :: D a) (ds :: Show b) a b
     = let dc :: D a = %p1 dd  -- Superclass selection
       in meth @a dc ....
          meth @a dc ....

When we specialise `f`, at a=Int say, that superclass selection can
fire (via rewiteClassOps), but that info (that 'dc' is now a
particular dictionary `C`, of type `C Int`) must be available to
the call `meth @a dc`, so that we can fire the `meth` class-op, and
thence specialise `wombat`.

We deliver on this idea by updating the unfolding for the binder
in the NonRec case of specBind.  (This is too exotic to trouble with
the Rec case.)
-}

interestingDict :: CoreExpr -> Type -> Bool
-- A dictionary argument is interesting if it has *some* structure,
-- see Note [Interesting dictionary arguments]
-- NB: "dictionary" arguments include constraints of all sorts,
--     including equality constraints; hence the Coercion case
-- To make this work, we need to ensure that dictionaries have
-- unfoldings in them.
interestingDict :: OutExpr -> Kind -> Bool
interestingDict OutExpr
arg Kind
arg_ty
  | Bool -> Bool
not (Kind -> Bool
typeDeterminesValue Kind
arg_ty) = Bool
False   -- See Note [Type determines value]
  | Bool
otherwise                        = OutExpr -> Bool
forall b. Expr b -> Bool
go OutExpr
arg
  where
    go :: Expr b -> Bool
go (Var Id
v)               =  Unfolding -> Bool
hasSomeUnfolding (Id -> Unfolding
idUnfolding Id
v)
                             Bool -> Bool -> Bool
|| Id -> Bool
isDataConWorkId Id
v
    go (Type Kind
_)              = Bool
False
    go (Coercion Coercion
_)          = Bool
False
    go (App Expr b
fn (Type Kind
_))     = Expr b -> Bool
go Expr b
fn
    go (App Expr b
fn (Coercion Coercion
_)) = Expr b -> Bool
go Expr b
fn
    go (Tick CoreTickish
_ Expr b
a)            = Expr b -> Bool
go Expr b
a
    go (Cast Expr b
e Coercion
_)            = Expr b -> Bool
go Expr b
e
    go Expr b
_                     = Bool
True

thenUDs :: UsageDetails -> UsageDetails -> UsageDetails
thenUDs :: UsageDetails -> UsageDetails -> UsageDetails
thenUDs (MkUD {ud_binds :: UsageDetails -> FloatedDictBinds
ud_binds = FloatedDictBinds
db1, ud_calls :: UsageDetails -> CallDetails
ud_calls = CallDetails
calls1})
        (MkUD {ud_binds :: UsageDetails -> FloatedDictBinds
ud_binds = FloatedDictBinds
db2, ud_calls :: UsageDetails -> CallDetails
ud_calls = CallDetails
calls2})
  = MkUD { ud_binds :: FloatedDictBinds
ud_binds       = FloatedDictBinds
db1    FloatedDictBinds -> FloatedDictBinds -> FloatedDictBinds
`thenFDBs`   FloatedDictBinds
db2
         , ud_calls :: CallDetails
ud_calls       = CallDetails
calls1 CallDetails -> CallDetails -> CallDetails
`unionCalls`  CallDetails
calls2 }

thenFDBs :: FloatedDictBinds -> FloatedDictBinds -> FloatedDictBinds
-- Combine FloatedDictBinds
-- In (dbs1 `thenFDBs` dbs2), dbs2 may mention dbs1 but not vice versa
thenFDBs :: FloatedDictBinds -> FloatedDictBinds -> FloatedDictBinds
thenFDBs (FDB { fdb_binds :: FloatedDictBinds -> OrdList DictBind
fdb_binds = OrdList DictBind
dbs1, fdb_bndrs :: FloatedDictBinds -> VarSet
fdb_bndrs = VarSet
bs1 })
         (FDB { fdb_binds :: FloatedDictBinds -> OrdList DictBind
fdb_binds = OrdList DictBind
dbs2, fdb_bndrs :: FloatedDictBinds -> VarSet
fdb_bndrs = VarSet
bs2 })
  = FDB { fdb_binds :: OrdList DictBind
fdb_binds = OrdList DictBind
dbs1 OrdList DictBind -> OrdList DictBind -> OrdList DictBind
forall a. OrdList a -> OrdList a -> OrdList a
`appOL` OrdList DictBind
dbs2
        , fdb_bndrs :: VarSet
fdb_bndrs = VarSet
bs1  VarSet -> VarSet -> VarSet
`unionVarSet` VarSet
bs2 }

-----------------------------
_dictBindBndrs :: OrdList DictBind -> [Id]
_dictBindBndrs :: OrdList DictBind -> [Id]
_dictBindBndrs OrdList DictBind
dbs = (DictBind -> [Id] -> [Id]) -> [Id] -> OrdList DictBind -> [Id]
forall a b. (a -> b -> b) -> b -> OrdList a -> b
forall (t :: * -> *) a b.
Foldable t =>
(a -> b -> b) -> b -> t a -> b
foldr ([Id] -> [Id] -> [Id]
forall a. [a] -> [a] -> [a]
(++) ([Id] -> [Id] -> [Id])
-> (DictBind -> [Id]) -> DictBind -> [Id] -> [Id]
forall b c a. (b -> c) -> (a -> b) -> a -> c
. InBind -> [Id]
forall b. Bind b -> [b]
bindersOf (InBind -> [Id]) -> (DictBind -> InBind) -> DictBind -> [Id]
forall b c a. (b -> c) -> (a -> b) -> a -> c
. DictBind -> InBind
db_bind) [] OrdList DictBind
dbs

-- | Construct a 'DictBind' from a 'CoreBind'
mkDB :: CoreBind -> DictBind
mkDB :: InBind -> DictBind
mkDB InBind
bind = DB { db_bind :: InBind
db_bind = InBind
bind, db_fvs :: VarSet
db_fvs = InBind -> VarSet
bind_fvs InBind
bind }

-- | Identify the free variables of a 'CoreBind'
bind_fvs :: CoreBind -> VarSet
bind_fvs :: InBind -> VarSet
bind_fvs (NonRec Id
bndr OutExpr
rhs) = (Id, OutExpr) -> VarSet
pair_fvs (Id
bndr,OutExpr
rhs)
bind_fvs (Rec [(Id, OutExpr)]
prs)         = VarSet
rhs_fvs VarSet -> [Id] -> VarSet
`delVarSetList` (((Id, OutExpr) -> Id) -> [(Id, OutExpr)] -> [Id]
forall a b. (a -> b) -> [a] -> [b]
map (Id, OutExpr) -> Id
forall a b. (a, b) -> a
fst [(Id, OutExpr)]
prs)
                           where
                             rhs_fvs :: VarSet
rhs_fvs = [VarSet] -> VarSet
unionVarSets (((Id, OutExpr) -> VarSet) -> [(Id, OutExpr)] -> [VarSet]
forall a b. (a -> b) -> [a] -> [b]
map (Id, OutExpr) -> VarSet
pair_fvs [(Id, OutExpr)]
prs)

pair_fvs :: (Id, CoreExpr) -> VarSet
pair_fvs :: (Id, OutExpr) -> VarSet
pair_fvs (Id
bndr, OutExpr
rhs) = (Id -> Bool) -> OutExpr -> VarSet
exprSomeFreeVars Id -> Bool
interesting OutExpr
rhs
                       VarSet -> VarSet -> VarSet
`unionVarSet` Id -> VarSet
idFreeVars Id
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 :: Id -> Bool
interesting Id
v = Id -> Bool
isLocalVar Id
v Bool -> Bool -> Bool
|| (Id -> Bool
isId Id
v Bool -> Bool -> Bool
&& Id -> Bool
isDFunId Id
v)
        -- Very important: include DFunIds /even/ if it is imported
        -- Reason: See Note [Avoiding loops in specImports], the #13429
        --         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

-- | Flatten a set of "dumped" 'DictBind's, and some other binding
-- pairs, into a single recursive binding.
recWithDumpedDicts :: [(Id,CoreExpr)] -> OrdList DictBind -> DictBind
recWithDumpedDicts :: [(Id, OutExpr)] -> OrdList DictBind -> DictBind
recWithDumpedDicts [(Id, OutExpr)]
pairs OrdList DictBind
dbs
  = DB { db_bind :: InBind
db_bind = [(Id, OutExpr)] -> InBind
forall b. [(b, Expr b)] -> Bind b
Rec [(Id, OutExpr)]
bindings
       , db_fvs :: VarSet
db_fvs = VarSet
fvs VarSet -> [Id] -> VarSet
`delVarSetList` ((Id, OutExpr) -> Id) -> [(Id, OutExpr)] -> [Id]
forall a b. (a -> b) -> [a] -> [b]
map (Id, OutExpr) -> Id
forall a b. (a, b) -> a
fst [(Id, OutExpr)]
bindings }
  where
    ([(Id, OutExpr)]
bindings, VarSet
fvs) = (DictBind
 -> ([(Id, OutExpr)], VarSet) -> ([(Id, OutExpr)], VarSet))
-> ([(Id, OutExpr)], VarSet)
-> OrdList DictBind
-> ([(Id, OutExpr)], VarSet)
forall a b. (a -> b -> b) -> b -> OrdList a -> b
forall (t :: * -> *) a b.
Foldable t =>
(a -> b -> b) -> b -> t a -> b
foldr DictBind -> ([(Id, OutExpr)], VarSet) -> ([(Id, OutExpr)], VarSet)
add ([], VarSet
emptyVarSet)
                                (OrdList DictBind
dbs OrdList DictBind -> DictBind -> OrdList DictBind
forall a. OrdList a -> a -> OrdList a
`snocOL` InBind -> DictBind
mkDB ([(Id, OutExpr)] -> InBind
forall b. [(b, Expr b)] -> Bind b
Rec [(Id, OutExpr)]
pairs))
    add :: DictBind -> ([(Id, OutExpr)], VarSet) -> ([(Id, OutExpr)], VarSet)
add (DB { db_bind :: DictBind -> InBind
db_bind = InBind
bind, db_fvs :: DictBind -> VarSet
db_fvs = VarSet
fvs }) ([(Id, OutExpr)]
prs_acc, VarSet
fvs_acc)
      = case InBind
bind of
          NonRec Id
b OutExpr
r -> ((Id
b,OutExpr
r) (Id, OutExpr) -> [(Id, OutExpr)] -> [(Id, OutExpr)]
forall a. a -> [a] -> [a]
: [(Id, OutExpr)]
prs_acc, VarSet
fvs')
          Rec [(Id, OutExpr)]
prs1   -> ([(Id, OutExpr)]
prs1 [(Id, OutExpr)] -> [(Id, OutExpr)] -> [(Id, OutExpr)]
forall a. [a] -> [a] -> [a]
++ [(Id, OutExpr)]
prs_acc, VarSet
fvs')
      where
        fvs' :: VarSet
fvs' = VarSet
fvs_acc VarSet -> VarSet -> VarSet
`unionVarSet` VarSet
fvs

snocDictBind :: UsageDetails -> DictBind -> UsageDetails
snocDictBind :: UsageDetails -> DictBind -> UsageDetails
snocDictBind uds :: UsageDetails
uds@MkUD{ud_binds :: UsageDetails -> FloatedDictBinds
ud_binds= FDB { fdb_binds :: FloatedDictBinds -> OrdList DictBind
fdb_binds = OrdList DictBind
dbs, fdb_bndrs :: FloatedDictBinds -> VarSet
fdb_bndrs = VarSet
bs }} DictBind
db
  = UsageDetails
uds { ud_binds = FDB { fdb_binds = dbs `snocOL` db
                         , fdb_bndrs = bs `extendVarSetList` bindersOfDictBind db } }

snocDictBinds :: UsageDetails -> [DictBind] -> UsageDetails
-- Add ud_binds to the tail end of the bindings in uds
snocDictBinds :: UsageDetails -> [DictBind] -> UsageDetails
snocDictBinds uds :: UsageDetails
uds@MkUD{ud_binds :: UsageDetails -> FloatedDictBinds
ud_binds=FDB{ fdb_binds :: FloatedDictBinds -> OrdList DictBind
fdb_binds = OrdList DictBind
binds, fdb_bndrs :: FloatedDictBinds -> VarSet
fdb_bndrs = VarSet
bs }} [DictBind]
dbs
  = UsageDetails
uds { ud_binds = FDB { fdb_binds = binds `appOL`        (toOL dbs)
                         , fdb_bndrs = bs    `extendVarSetList` bindersOfDictBinds dbs } }

consDictBinds :: [DictBind] -> UsageDetails -> UsageDetails
consDictBinds :: [DictBind] -> UsageDetails -> UsageDetails
consDictBinds [DictBind]
dbs uds :: UsageDetails
uds@MkUD{ud_binds :: UsageDetails -> FloatedDictBinds
ud_binds=FDB{fdb_binds :: FloatedDictBinds -> OrdList DictBind
fdb_binds = OrdList DictBind
binds, fdb_bndrs :: FloatedDictBinds -> VarSet
fdb_bndrs = VarSet
bs}}
  = UsageDetails
uds { ud_binds = FDB{ fdb_binds = toOL dbs `appOL` binds
                        , fdb_bndrs = bs `extendVarSetList` bindersOfDictBinds dbs } }

wrapDictBinds :: FloatedDictBinds -> [CoreBind] -> [CoreBind]
wrapDictBinds :: FloatedDictBinds -> CoreProgram -> CoreProgram
wrapDictBinds (FDB { fdb_binds :: FloatedDictBinds -> OrdList DictBind
fdb_binds = OrdList DictBind
dbs }) CoreProgram
binds
  = (DictBind -> CoreProgram -> CoreProgram)
-> CoreProgram -> OrdList DictBind -> CoreProgram
forall a b. (a -> b -> b) -> b -> OrdList a -> b
forall (t :: * -> *) a b.
Foldable t =>
(a -> b -> b) -> b -> t a -> b
foldr DictBind -> CoreProgram -> CoreProgram
add CoreProgram
binds OrdList DictBind
dbs
  where
    add :: DictBind -> CoreProgram -> CoreProgram
add (DB { db_bind :: DictBind -> InBind
db_bind = InBind
bind }) CoreProgram
binds = InBind
bind InBind -> CoreProgram -> CoreProgram
forall a. a -> [a] -> [a]
: CoreProgram
binds

wrapDictBindsE :: OrdList DictBind -> CoreExpr -> CoreExpr
wrapDictBindsE :: OrdList DictBind -> OutExpr -> OutExpr
wrapDictBindsE OrdList DictBind
dbs OutExpr
expr
  = (DictBind -> OutExpr -> OutExpr)
-> OutExpr -> OrdList DictBind -> OutExpr
forall a b. (a -> b -> b) -> b -> OrdList a -> b
forall (t :: * -> *) a b.
Foldable t =>
(a -> b -> b) -> b -> t a -> b
foldr DictBind -> OutExpr -> OutExpr
add OutExpr
expr OrdList DictBind
dbs
  where
    add :: DictBind -> OutExpr -> OutExpr
add (DB { db_bind :: DictBind -> InBind
db_bind = InBind
bind }) OutExpr
expr = InBind -> OutExpr -> OutExpr
forall b. Bind b -> Expr b -> Expr b
Let InBind
bind OutExpr
expr

----------------------
dumpUDs :: [CoreBndr] -> UsageDetails -> (UsageDetails, OrdList DictBind)
-- Used at a lambda or case binder; just dump anything mentioning the binder
dumpUDs :: [Id] -> UsageDetails -> (UsageDetails, OrdList DictBind)
dumpUDs [Id]
bndrs uds :: UsageDetails
uds@(MkUD { ud_binds :: UsageDetails -> FloatedDictBinds
ud_binds = FloatedDictBinds
orig_dbs, ud_calls :: UsageDetails -> CallDetails
ud_calls = CallDetails
orig_calls })
  | [Id] -> Bool
forall a. [a] -> Bool
forall (t :: * -> *) a. Foldable t => t a -> Bool
null [Id]
bndrs = (UsageDetails
uds, OrdList DictBind
forall a. OrdList a
nilOL)  -- Common in case alternatives
  | Bool
otherwise  = -- pprTrace "dumpUDs" (vcat
                 --    [ text "bndrs" <+> ppr bndrs
                 --    , text "uds" <+> ppr uds
                 --    , text "free_uds" <+> ppr free_uds
                 --    , text "dump-dbs" <+> ppr dump_dbs ]) $
                 (UsageDetails
free_uds, OrdList DictBind
dump_dbs)
  where
    free_uds :: UsageDetails
free_uds = UsageDetails
uds { ud_binds = free_dbs, ud_calls = free_calls }
    bndr_set :: VarSet
bndr_set = [Id] -> VarSet
mkVarSet [Id]
bndrs
    (FloatedDictBinds
free_dbs, OrdList DictBind
dump_dbs, VarSet
dump_set) = FloatedDictBinds
-> VarSet -> (FloatedDictBinds, OrdList DictBind, VarSet)
splitDictBinds FloatedDictBinds
orig_dbs VarSet
bndr_set
    free_calls :: CallDetails
free_calls = VarSet -> CallDetails -> CallDetails
deleteCallsMentioning VarSet
dump_set (CallDetails -> CallDetails) -> CallDetails -> CallDetails
forall a b. (a -> b) -> a -> b
$   -- Drop calls mentioning bndr_set on the floor
                 [Id] -> CallDetails -> CallDetails
deleteCallsFor [Id]
bndrs CallDetails
orig_calls    -- Discard calls for bndr_set; there should be
                                                    -- no calls for any of the dicts in dump_dbs

dumpBindUDs :: [CoreBndr] -> UsageDetails -> (UsageDetails, OrdList 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 :: [Id] -> UsageDetails -> (UsageDetails, OrdList DictBind, Bool)
dumpBindUDs [Id]
bndrs (MkUD { ud_binds :: UsageDetails -> FloatedDictBinds
ud_binds = FloatedDictBinds
orig_dbs, ud_calls :: UsageDetails -> CallDetails
ud_calls = CallDetails
orig_calls })
  = -- pprTrace "dumpBindUDs" (ppr bndrs $$ ppr free_uds $$ ppr dump_dbs $$ ppr float_all) $
    (UsageDetails
free_uds, OrdList DictBind
dump_dbs, Bool
float_all)
  where
    free_uds :: UsageDetails
free_uds = MkUD { ud_binds :: FloatedDictBinds
ud_binds = FloatedDictBinds
free_dbs, ud_calls :: CallDetails
ud_calls = CallDetails
free_calls }
    bndr_set :: VarSet
bndr_set = [Id] -> VarSet
mkVarSet [Id]
bndrs
    (FloatedDictBinds
free_dbs, OrdList DictBind
dump_dbs, VarSet
dump_set) = FloatedDictBinds
-> VarSet -> (FloatedDictBinds, OrdList DictBind, VarSet)
splitDictBinds FloatedDictBinds
orig_dbs VarSet
bndr_set
    free_calls :: CallDetails
free_calls = [Id] -> CallDetails -> CallDetails
deleteCallsFor [Id]
bndrs CallDetails
orig_calls
    float_all :: Bool
float_all = VarSet
dump_set VarSet -> VarSet -> Bool
`intersectsVarSet` CallDetails -> VarSet
callDetailsFVs CallDetails
free_calls

callsForMe :: Id -> UsageDetails -> (UsageDetails, [CallInfo])
callsForMe :: Id -> UsageDetails -> (UsageDetails, [CallInfo])
callsForMe Id
fn uds :: UsageDetails
uds@MkUD { ud_binds :: UsageDetails -> FloatedDictBinds
ud_binds = FloatedDictBinds
orig_dbs, ud_calls :: UsageDetails -> CallDetails
ud_calls = CallDetails
orig_calls }
  = -- pprTrace ("callsForMe")
    --          (vcat [ppr fn,
    --                 text "Orig dbs ="     <+> ppr (_dictBindBndrs orig_dbs),
    --                 text "Orig calls ="   <+> ppr orig_calls,
    --                 text "Calls for me =" <+> ppr calls_for_me]) $
    (UsageDetails
uds_without_me, [CallInfo]
calls_for_me)
  where
    uds_without_me :: UsageDetails
uds_without_me = UsageDetails
uds { ud_calls = delDVarEnv orig_calls fn }
    calls_for_me :: [CallInfo]
calls_for_me = case CallDetails -> Id -> Maybe CallInfoSet
forall a. DVarEnv a -> Id -> Maybe a
lookupDVarEnv CallDetails
orig_calls Id
fn of
                        Maybe CallInfoSet
Nothing -> []
                        Just CallInfoSet
cis -> CallInfoSet -> FloatedDictBinds -> [CallInfo]
filterCalls CallInfoSet
cis FloatedDictBinds
orig_dbs

----------------------
filterCalls :: CallInfoSet -> FloatedDictBinds -> [CallInfo]
-- Remove
--   (a) dominated calls: (MP3) in Note [Specialising polymorphic dictionaries]
--   (b) loopy DFuns: Note [Avoiding loops (DFuns)]
filterCalls :: CallInfoSet -> FloatedDictBinds -> [CallInfo]
filterCalls (CIS Id
fn Bag CallInfo
call_bag) (FDB { fdb_binds :: FloatedDictBinds -> OrdList DictBind
fdb_binds = OrdList DictBind
dbs })
  | Id -> Bool
isDFunId Id
fn  = (CallInfo -> Bool) -> [CallInfo] -> [CallInfo]
forall a. (a -> Bool) -> [a] -> [a]
filter CallInfo -> Bool
ok_call [CallInfo]
de_dupd_calls  -- Deals with (b)
  | Bool
otherwise    = [CallInfo]
de_dupd_calls
  where
    de_dupd_calls :: [CallInfo]
de_dupd_calls = Bag CallInfo -> [CallInfo]
removeDupCalls Bag CallInfo
call_bag -- Deals with (a)

    dump_set :: VarSet
dump_set = (VarSet -> DictBind -> VarSet)
-> VarSet -> OrdList DictBind -> VarSet
forall b a. (b -> a -> b) -> b -> OrdList a -> b
forall (t :: * -> *) b a.
Foldable t =>
(b -> a -> b) -> b -> t a -> b
foldl' VarSet -> DictBind -> VarSet
go (Id -> VarSet
unitVarSet Id
fn) OrdList DictBind
dbs
      -- This dump-set could also be computed by splitDictBinds
      --   (_,_,dump_set) = splitDictBinds dbs {fn}
      -- But this variant is shorter

    go :: VarSet -> DictBind -> VarSet
go VarSet
so_far (DB { db_bind :: DictBind -> InBind
db_bind = InBind
bind, db_fvs :: DictBind -> VarSet
db_fvs = VarSet
fvs })
       | VarSet
fvs VarSet -> VarSet -> Bool
`intersectsVarSet` VarSet
so_far
       = VarSet -> [Id] -> VarSet
extendVarSetList VarSet
so_far (InBind -> [Id]
forall b. Bind b -> [b]
bindersOf InBind
bind)
       | Bool
otherwise = VarSet
so_far

    ok_call :: CallInfo -> Bool
ok_call (CI { ci_fvs :: CallInfo -> VarSet
ci_fvs = VarSet
fvs }) = VarSet
fvs VarSet -> VarSet -> Bool
`disjointVarSet` VarSet
dump_set

removeDupCalls :: Bag CallInfo -> [CallInfo]
-- Calls involving more generic instances beat more specific ones.
-- See (MP3) in Note [Specialising polymorphic dictionaries]
removeDupCalls :: Bag CallInfo -> [CallInfo]
removeDupCalls Bag CallInfo
calls = (CallInfo -> [CallInfo] -> [CallInfo])
-> [CallInfo] -> Bag CallInfo -> [CallInfo]
forall a b. (a -> b -> b) -> b -> Bag a -> b
forall (t :: * -> *) a b.
Foldable t =>
(a -> b -> b) -> b -> t a -> b
foldr CallInfo -> [CallInfo] -> [CallInfo]
add [] Bag CallInfo
calls
  where
    add :: CallInfo -> [CallInfo] -> [CallInfo]
    add :: CallInfo -> [CallInfo] -> [CallInfo]
add CallInfo
ci [] = [CallInfo
ci]
    add CallInfo
ci1 (CallInfo
ci2:[CallInfo]
cis) | CallInfo
ci2 CallInfo -> CallInfo -> Bool
`beats_or_same` CallInfo
ci1 = CallInfo
ci2CallInfo -> [CallInfo] -> [CallInfo]
forall a. a -> [a] -> [a]
:[CallInfo]
cis
                      | CallInfo
ci1 CallInfo -> CallInfo -> Bool
`beats_or_same` CallInfo
ci2 = CallInfo
ci1CallInfo -> [CallInfo] -> [CallInfo]
forall a. a -> [a] -> [a]
:[CallInfo]
cis
                      | Bool
otherwise               = CallInfo
ci2 CallInfo -> [CallInfo] -> [CallInfo]
forall a. a -> [a] -> [a]
: CallInfo -> [CallInfo] -> [CallInfo]
add CallInfo
ci1 [CallInfo]
cis

beats_or_same :: CallInfo -> CallInfo -> Bool
-- (beats_or_same ci1 ci2) is True if specialising on ci1 subsumes ci2
-- That is: ci1's types are less specialised than ci2
--          ci1   specialises on the same dict args as ci2
beats_or_same :: CallInfo -> CallInfo -> Bool
beats_or_same (CI { ci_key :: CallInfo -> [SpecArg]
ci_key = [SpecArg]
args1 }) (CI { ci_key :: CallInfo -> [SpecArg]
ci_key = [SpecArg]
args2 })
  = [SpecArg] -> [SpecArg] -> Bool
go [SpecArg]
args1 [SpecArg]
args2
  where
    go :: [SpecArg] -> [SpecArg] -> Bool
go []           []           = Bool
True
    go (SpecArg
arg1:[SpecArg]
args1) (SpecArg
arg2:[SpecArg]
args2) = SpecArg -> SpecArg -> Bool
go_arg SpecArg
arg1 SpecArg
arg2 Bool -> Bool -> Bool
&& [SpecArg] -> [SpecArg] -> Bool
go [SpecArg]
args1 [SpecArg]
args2

    -- If one or the other runs dry, the other must still have a SpecDict
    -- because of the (CI-KEY) invariant.  So neither subsumes the other;
    -- one is more specialised (faster code) but the other is more generally
    -- applicable.
    go  [SpecArg]
_ [SpecArg]
_ = Bool
False

    go_arg :: SpecArg -> SpecArg -> Bool
go_arg (SpecType Kind
ty1) (SpecType Kind
ty2) = Maybe Subst -> Bool
forall a. Maybe a -> Bool
isJust (HasDebugCallStack => Kind -> Kind -> Maybe Subst
Kind -> Kind -> Maybe Subst
tcMatchTy Kind
ty1 Kind
ty2)
    go_arg SpecArg
UnspecType     SpecArg
UnspecType     = Bool
True
    go_arg (SpecDict {})  (SpecDict {})  = Bool
True
    go_arg SpecArg
UnspecArg      SpecArg
UnspecArg      = Bool
True
    go_arg SpecArg
_              SpecArg
_              = Bool
False

----------------------
splitDictBinds :: FloatedDictBinds -> IdSet -> (FloatedDictBinds, OrdList 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 :: FloatedDictBinds
-> VarSet -> (FloatedDictBinds, OrdList DictBind, VarSet)
splitDictBinds (FDB { fdb_binds :: FloatedDictBinds -> OrdList DictBind
fdb_binds = OrdList DictBind
dbs, fdb_bndrs :: FloatedDictBinds -> VarSet
fdb_bndrs = VarSet
bs }) VarSet
bndr_set
   = (FDB { fdb_binds :: OrdList DictBind
fdb_binds = OrdList DictBind
free_dbs
          , fdb_bndrs :: VarSet
fdb_bndrs = VarSet
bs VarSet -> VarSet -> VarSet
`minusVarSet` VarSet
dump_set }
     , OrdList DictBind
dump_dbs, VarSet
dump_set)
   where
    (OrdList DictBind
free_dbs, OrdList DictBind
dump_dbs, VarSet
dump_set)
      = ((OrdList DictBind, OrdList DictBind, VarSet)
 -> DictBind -> (OrdList DictBind, OrdList DictBind, VarSet))
-> (OrdList DictBind, OrdList DictBind, VarSet)
-> OrdList DictBind
-> (OrdList DictBind, OrdList DictBind, VarSet)
forall b a. (b -> a -> b) -> b -> OrdList a -> b
forall (t :: * -> *) b a.
Foldable t =>
(b -> a -> b) -> b -> t a -> b
foldl' (OrdList DictBind, OrdList DictBind, VarSet)
-> DictBind -> (OrdList DictBind, OrdList DictBind, VarSet)
split_db (OrdList DictBind
forall a. OrdList a
nilOL, OrdList DictBind
forall a. OrdList a
nilOL, VarSet
bndr_set) OrdList DictBind
dbs
                -- Important that it's foldl' not foldr;
                -- we're accumulating the set of dumped ids in dump_set

    split_db :: (OrdList DictBind, OrdList DictBind, VarSet)
-> DictBind -> (OrdList DictBind, OrdList DictBind, VarSet)
split_db (OrdList DictBind
free_dbs, OrdList DictBind
dump_dbs, VarSet
dump_idset) DictBind
db
        | DB { db_bind :: DictBind -> InBind
db_bind = InBind
bind, db_fvs :: DictBind -> VarSet
db_fvs = VarSet
fvs } <- DictBind
db
        , VarSet
dump_idset VarSet -> VarSet -> Bool
`intersectsVarSet` VarSet
fvs     -- Dump it
        = (OrdList DictBind
free_dbs, OrdList DictBind
dump_dbs OrdList DictBind -> DictBind -> OrdList DictBind
forall a. OrdList a -> a -> OrdList a
`snocOL` DictBind
db,
           VarSet -> [Id] -> VarSet
extendVarSetList VarSet
dump_idset (InBind -> [Id]
forall b. Bind b -> [b]
bindersOf InBind
bind))

        | Bool
otherwise     -- Don't dump it
        = (OrdList DictBind
free_dbs OrdList DictBind -> DictBind -> OrdList DictBind
forall a. OrdList a -> a -> OrdList a
`snocOL` DictBind
db, OrdList DictBind
dump_dbs, VarSet
dump_idset)


----------------------
deleteCallsMentioning :: VarSet -> CallDetails -> CallDetails
-- Remove calls mentioning any Id in bndrs
-- NB: The call is allowed to mention TyVars in bndrs
--     Note [Specialising polymorphic dictionaries]
--     ci_fvs are just the free /Ids/
deleteCallsMentioning :: VarSet -> CallDetails -> CallDetails
deleteCallsMentioning VarSet
bndrs CallDetails
calls
  = (CallInfoSet -> CallInfoSet) -> CallDetails -> CallDetails
forall a b. (a -> b) -> DVarEnv a -> DVarEnv b
mapDVarEnv ((CallInfo -> Bool) -> CallInfoSet -> CallInfoSet
ciSetFilter CallInfo -> Bool
keep_call) CallDetails
calls
  where
    keep_call :: CallInfo -> Bool
keep_call (CI { ci_fvs :: CallInfo -> VarSet
ci_fvs = VarSet
fvs }) = VarSet
fvs VarSet -> VarSet -> Bool
`disjointVarSet` VarSet
bndrs

deleteCallsFor :: [Id] -> CallDetails -> CallDetails
-- Remove calls *for* bndrs
deleteCallsFor :: [Id] -> CallDetails -> CallDetails
deleteCallsFor [Id]
bndrs CallDetails
calls = CallDetails -> [Id] -> CallDetails
forall a. DVarEnv a -> [Id] -> DVarEnv a
delDVarEnvList CallDetails
calls [Id]
bndrs

{-
************************************************************************
*                                                                      *
\subsubsection{Boring helper functions}
*                                                                      *
************************************************************************
-}

type SpecM a = UniqSM a

runSpecM :: SpecM a -> CoreM a
runSpecM :: forall a. SpecM a -> CoreM a
runSpecM SpecM a
thing_inside
  = do { us <- CoreM UniqSupply
forall (m :: * -> *). MonadUnique m => m UniqSupply
getUniqueSupplyM
       ; return (initUs_ us thing_inside) }

mapAndCombineSM :: (a -> SpecM (b, UsageDetails)) -> [a] -> SpecM ([b], UsageDetails)
mapAndCombineSM :: forall a b.
(a -> SpecM (b, UsageDetails)) -> [a] -> SpecM ([b], UsageDetails)
mapAndCombineSM a -> SpecM (b, UsageDetails)
_ []     = ([b], UsageDetails) -> UniqSM ([b], UsageDetails)
forall a. a -> UniqSM a
forall (m :: * -> *) a. Monad m => a -> m a
return ([], UsageDetails
emptyUDs)
mapAndCombineSM a -> SpecM (b, UsageDetails)
f (a
x:[a]
xs) = do (y, uds1) <- a -> SpecM (b, UsageDetails)
f a
x
                              (ys, uds2) <- mapAndCombineSM f xs
                              return (y:ys, uds1 `thenUDs` uds2)

extendTvSubst :: SpecEnv -> TyVar -> Type -> SpecEnv
extendTvSubst :: SpecEnv -> Id -> Kind -> SpecEnv
extendTvSubst SpecEnv
env Id
tv Kind
ty
  = SpecEnv
env { se_subst = Core.extendTvSubst (se_subst env) tv ty }

extendInScope :: SpecEnv -> OutId -> SpecEnv
extendInScope :: SpecEnv -> Id -> SpecEnv
extendInScope env :: SpecEnv
env@(SE { se_subst :: SpecEnv -> Subst
se_subst = Subst
subst }) Id
bndr
  = SpecEnv
env { se_subst = subst `Core.extendSubstInScope` bndr }

zapSubst :: SpecEnv -> SpecEnv
zapSubst :: SpecEnv -> SpecEnv
zapSubst env :: SpecEnv
env@(SE { se_subst :: SpecEnv -> Subst
se_subst = Subst
subst })
  = SpecEnv
env { se_subst = Core.zapSubst subst }

substTy :: SpecEnv -> Type -> Type
substTy :: SpecEnv -> Kind -> Kind
substTy SpecEnv
env Kind
ty = Subst -> Kind -> Kind
substTyUnchecked (SpecEnv -> Subst
se_subst SpecEnv
env) Kind
ty

substCo :: SpecEnv -> Coercion -> Coercion
substCo :: SpecEnv -> Coercion -> Coercion
substCo SpecEnv
env Coercion
co = HasDebugCallStack => Subst -> Coercion -> Coercion
Subst -> Coercion -> Coercion
Core.substCo (SpecEnv -> Subst
se_subst SpecEnv
env) Coercion
co

substBndr :: SpecEnv -> CoreBndr -> (SpecEnv, CoreBndr)
substBndr :: SpecEnv -> Id -> (SpecEnv, Id)
substBndr SpecEnv
env Id
bs = case Subst -> Id -> (Subst, Id)
Core.substBndr (SpecEnv -> Subst
se_subst SpecEnv
env) Id
bs of
                      (Subst
subst', Id
bs') -> (SpecEnv
env { se_subst = subst' }, Id
bs')

substBndrs :: Traversable f => SpecEnv -> f CoreBndr -> (SpecEnv, f CoreBndr)
substBndrs :: forall (f :: * -> *).
Traversable f =>
SpecEnv -> f Id -> (SpecEnv, f Id)
substBndrs SpecEnv
env f Id
bs = case Subst -> f Id -> (Subst, f Id)
forall (f :: * -> *).
Traversable f =>
Subst -> f Id -> (Subst, f Id)
Core.substBndrs (SpecEnv -> Subst
se_subst SpecEnv
env) f Id
bs of
                      (Subst
subst', f Id
bs') -> (SpecEnv
env { se_subst = subst' }, f Id
bs')

cloneBndrSM :: SpecEnv -> Id -> SpecM (SpecEnv, Id)
-- 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
-- Discards non-Stable unfoldings
cloneBndrSM :: SpecEnv -> Id -> UniqSM (SpecEnv, Id)
cloneBndrSM env :: SpecEnv
env@(SE { se_subst :: SpecEnv -> Subst
se_subst = Subst
subst }) Id
bndr
  = do { us <- UniqSM UniqSupply
forall (m :: * -> *). MonadUnique m => m UniqSupply
getUniqueSupplyM
       ; let (subst', bndr') = Core.cloneIdBndr subst us bndr
       ; return (env { se_subst = subst' }, bndr') }

cloneRecBndrsSM :: SpecEnv -> [Id] -> SpecM (SpecEnv, [Id])
cloneRecBndrsSM :: SpecEnv -> [Id] -> UniqSM (SpecEnv, [Id])
cloneRecBndrsSM env :: SpecEnv
env@(SE { se_subst :: SpecEnv -> Subst
se_subst = Subst
subst }) [Id]
bndrs
  = do { (subst', bndrs') <- Subst -> [Id] -> UniqSM (Subst, [Id])
forall (m :: * -> *).
MonadUnique m =>
Subst -> [Id] -> m (Subst, [Id])
Core.cloneRecIdBndrsM Subst
subst [Id]
bndrs
       ; let env' = SpecEnv
env { se_subst = subst' }
       ; return (env', bndrs') }

newDictBndr :: SpecEnv -> CoreBndr -> SpecM (SpecEnv, CoreBndr)
-- Make up completely fresh binders for the dictionaries
-- Their bindings are going to float outwards
newDictBndr :: SpecEnv -> Id -> UniqSM (SpecEnv, Id)
newDictBndr env :: SpecEnv
env@(SE { se_subst :: SpecEnv -> Subst
se_subst = Subst
subst }) Id
b
  = do { uniq <- UniqSM Unique
forall (m :: * -> *). MonadUnique m => m Unique
getUniqueM
       ; let n    = Id -> Name
idName Id
b
             ty'  = Subst -> Kind -> Kind
substTyUnchecked Subst
subst (Id -> Kind
idType Id
b)
             b'   = OccName -> Unique -> Kind -> Kind -> SrcSpan -> Id
mkUserLocal (Name -> OccName
nameOccName Name
n) Unique
uniq Kind
ManyTy Kind
ty' (Name -> SrcSpan
forall a. NamedThing a => a -> SrcSpan
getSrcSpan Name
n)
             env' = SpecEnv
env { se_subst = subst `Core.extendSubstInScope` b' }
       ; pure (env', b') }

newSpecIdSM :: Name -> Type -> IdDetails -> IdInfo -> SpecM Id
    -- Give the new Id a similar occurrence name to the old one
newSpecIdSM :: Name -> Kind -> IdDetails -> IdInfo -> UniqSM Id
newSpecIdSM Name
old_name Kind
new_ty IdDetails
details IdInfo
info
  = do  { uniq <- UniqSM Unique
forall (m :: * -> *). MonadUnique m => m Unique
getUniqueM
        ; let new_occ  = OccName -> OccName
mkSpecOcc (Name -> OccName
nameOccName Name
old_name)
              new_name = Unique -> OccName -> SrcSpan -> Name
mkInternalName Unique
uniq OccName
new_occ  (Name -> SrcSpan
forall a. NamedThing a => a -> SrcSpan
getSrcSpan Name
old_name)
        ; return (assert (not (isCoVarType new_ty)) $
                  mkLocalVar details new_name ManyTy new_ty info) }

{-
                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.
-}