{- (c) The GRASP/AQUA Project, Glasgow University, 1992-1998 \section[PrimOp]{Primitive operations (machine-level)} -} {-# LANGUAGE CPP #-} module PrimOp ( PrimOp(..), PrimOpVecCat(..), allThePrimOps, primOpType, primOpSig, primOpTag, maxPrimOpTag, primOpOcc, tagToEnumKey, primOpOutOfLine, primOpCodeSize, primOpOkForSpeculation, primOpOkForSideEffects, primOpIsCheap, primOpFixity, getPrimOpResultInfo, PrimOpResultInfo(..), PrimCall(..) ) where #include "HsVersions.h" import TysPrim import TysWiredIn import CmmType import Demand import Var ( TyVar ) import OccName ( OccName, pprOccName, mkVarOccFS ) import TyCon ( TyCon, isPrimTyCon, tyConPrimRep, PrimRep(..) ) import Type ( Type, mkForAllTys, mkFunTy, mkFunTys, tyConAppTyCon, typePrimRep ) import BasicTypes ( Arity, Fixity(..), FixityDirection(..), TupleSort(..) ) import ForeignCall ( CLabelString ) import Unique ( Unique, mkPrimOpIdUnique ) import Outputable import FastTypes import FastString import Module ( PackageKey ) {- ************************************************************************ * * \subsection[PrimOp-datatype]{Datatype for @PrimOp@ (an enumeration)} * * ************************************************************************ These are in \tr{state-interface.verb} order. -} -- supplies: -- data PrimOp = ... #include "primop-data-decl.hs-incl" -- Used for the Ord instance primOpTag :: PrimOp -> Int primOpTag op = iBox (tagOf_PrimOp op) -- supplies -- tagOf_PrimOp :: PrimOp -> FastInt #include "primop-tag.hs-incl" tagOf_PrimOp _ = error "tagOf_PrimOp: unknown primop" instance Eq PrimOp where op1 == op2 = tagOf_PrimOp op1 ==# tagOf_PrimOp op2 instance Ord PrimOp where op1 < op2 = tagOf_PrimOp op1 <# tagOf_PrimOp op2 op1 <= op2 = tagOf_PrimOp op1 <=# tagOf_PrimOp op2 op1 >= op2 = tagOf_PrimOp op1 >=# tagOf_PrimOp op2 op1 > op2 = tagOf_PrimOp op1 ># tagOf_PrimOp op2 op1 `compare` op2 | op1 < op2 = LT | op1 == op2 = EQ | otherwise = GT instance Outputable PrimOp where ppr op = pprPrimOp op data PrimOpVecCat = IntVec | WordVec | FloatVec -- An @Enum@-derived list would be better; meanwhile... (ToDo) allThePrimOps :: [PrimOp] allThePrimOps = #include "primop-list.hs-incl" tagToEnumKey :: Unique tagToEnumKey = mkPrimOpIdUnique (primOpTag TagToEnumOp) {- ************************************************************************ * * \subsection[PrimOp-info]{The essential info about each @PrimOp@} * * ************************************************************************ The @String@ in the @PrimOpInfos@ is the ``base name'' by which the user may refer to the primitive operation. The conventional \tr{#}-for- unboxed ops is added on later. The reason for the funny characters in the names is so we do not interfere with the programmer's Haskell name spaces. We use @PrimKinds@ for the ``type'' information, because they're (slightly) more convenient to use than @TyCons@. -} data PrimOpInfo = Dyadic OccName -- string :: T -> T -> T Type | Monadic OccName -- string :: T -> T Type | Compare OccName -- string :: T -> T -> Int# Type | GenPrimOp OccName -- string :: \/a1..an . T1 -> .. -> Tk -> T [TyVar] [Type] Type mkDyadic, mkMonadic, mkCompare :: FastString -> Type -> PrimOpInfo mkDyadic str ty = Dyadic (mkVarOccFS str) ty mkMonadic str ty = Monadic (mkVarOccFS str) ty mkCompare str ty = Compare (mkVarOccFS str) ty mkGenPrimOp :: FastString -> [TyVar] -> [Type] -> Type -> PrimOpInfo mkGenPrimOp str tvs tys ty = GenPrimOp (mkVarOccFS str) tvs tys ty {- ************************************************************************ * * \subsubsection{Strictness} * * ************************************************************************ Not all primops are strict! -} primOpStrictness :: PrimOp -> Arity -> StrictSig -- See Demand.StrictnessInfo for discussion of what the results -- The arity should be the arity of the primop; that's why -- this function isn't exported. #include "primop-strictness.hs-incl" {- ************************************************************************ * * \subsubsection{Fixity} * * ************************************************************************ -} primOpFixity :: PrimOp -> Maybe Fixity #include "primop-fixity.hs-incl" {- ************************************************************************ * * \subsubsection[PrimOp-comparison]{PrimOpInfo basic comparison ops} * * ************************************************************************ @primOpInfo@ gives all essential information (from which everything else, notably a type, can be constructed) for each @PrimOp@. -} primOpInfo :: PrimOp -> PrimOpInfo #include "primop-primop-info.hs-incl" primOpInfo _ = error "primOpInfo: unknown primop" {- Here are a load of comments from the old primOp info: A @Word#@ is an unsigned @Int#@. @decodeFloat#@ is given w/ Integer-stuff (it's similar). @decodeDouble#@ is given w/ Integer-stuff (it's similar). Decoding of floating-point numbers is sorta Integer-related. Encoding is done with plain ccalls now (see PrelNumExtra.lhs). A @Weak@ Pointer is created by the @mkWeak#@ primitive: mkWeak# :: k -> v -> f -> State# RealWorld -> (# State# RealWorld, Weak# v #) In practice, you'll use the higher-level data Weak v = Weak# v mkWeak :: k -> v -> IO () -> IO (Weak v) The following operation dereferences a weak pointer. The weak pointer may have been finalized, so the operation returns a result code which must be inspected before looking at the dereferenced value. deRefWeak# :: Weak# v -> State# RealWorld -> (# State# RealWorld, v, Int# #) Only look at v if the Int# returned is /= 0 !! The higher-level op is deRefWeak :: Weak v -> IO (Maybe v) Weak pointers can be finalized early by using the finalize# operation: finalizeWeak# :: Weak# v -> State# RealWorld -> (# State# RealWorld, Int#, IO () #) The Int# returned is either 0 if the weak pointer has already been finalized, or it has no finalizer (the third component is then invalid). 1 if the weak pointer is still alive, with the finalizer returned as the third component. A {\em stable name/pointer} is an index into a table of stable name entries. Since the garbage collector is told about stable pointers, it is safe to pass a stable pointer to external systems such as C routines. \begin{verbatim} makeStablePtr# :: a -> State# RealWorld -> (# State# RealWorld, StablePtr# a #) freeStablePtr :: StablePtr# a -> State# RealWorld -> State# RealWorld deRefStablePtr# :: StablePtr# a -> State# RealWorld -> (# State# RealWorld, a #) eqStablePtr# :: StablePtr# a -> StablePtr# a -> Int# \end{verbatim} It may seem a bit surprising that @makeStablePtr#@ is a @IO@ operation since it doesn't (directly) involve IO operations. The reason is that if some optimisation pass decided to duplicate calls to @makeStablePtr#@ and we only pass one of the stable pointers over, a massive space leak can result. Putting it into the IO monad prevents this. (Another reason for putting them in a monad is to ensure correct sequencing wrt the side-effecting @freeStablePtr@ operation.) An important property of stable pointers is that if you call makeStablePtr# twice on the same object you get the same stable pointer back. Note that we can implement @freeStablePtr#@ using @_ccall_@ (and, besides, it's not likely to be used from Haskell) so it's not a primop. Question: Why @RealWorld@ - won't any instance of @_ST@ do the job? [ADR] Stable Names ~~~~~~~~~~~~ A stable name is like a stable pointer, but with three important differences: (a) You can't deRef one to get back to the original object. (b) You can convert one to an Int. (c) You don't need to 'freeStableName' The existence of a stable name doesn't guarantee to keep the object it points to alive (unlike a stable pointer), hence (a). Invariants: (a) makeStableName always returns the same value for a given object (same as stable pointers). (b) if two stable names are equal, it implies that the objects from which they were created were the same. (c) stableNameToInt always returns the same Int for a given stable name. -- HWL: The first 4 Int# in all par... annotations denote: -- name, granularity info, size of result, degree of parallelism -- Same structure as _seq_ i.e. returns Int# -- KSW: v, the second arg in parAt# and parAtForNow#, is used only to determine -- `the processor containing the expression v'; it is not evaluated These primops are pretty weird. dataToTag# :: a -> Int (arg must be an evaluated data type) tagToEnum# :: Int -> a (result type must be an enumerated type) The constraints aren't currently checked by the front end, but the code generator will fall over if they aren't satisfied. ************************************************************************ * * Which PrimOps are out-of-line * * ************************************************************************ Some PrimOps need to be called out-of-line because they either need to perform a heap check or they block. -} primOpOutOfLine :: PrimOp -> Bool #include "primop-out-of-line.hs-incl" {- ************************************************************************ * * Failure and side effects * * ************************************************************************ Note [PrimOp can_fail and has_side_effects] ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Both can_fail and has_side_effects mean that the primop has some effect that is not captured entirely by its result value. ---------- has_side_effects --------------------- A primop "has_side_effects" if it has some *write* effect, visible elsewhere - writing to the world (I/O) - writing to a mutable data structure (writeIORef) - throwing a synchronous Haskell exception Often such primops have a type like State -> input -> (State, output) so the state token guarantees ordering. In general we rely *only* on data dependencies of the state token to enforce write-effect ordering * NB1: if you inline unsafePerformIO, you may end up with side-effecting ops whose 'state' output is discarded. And programmers may do that by hand; see Trac #9390. That is why we (conservatively) do not discard write-effecting primops even if both their state and result is discarded. * NB2: We consider primops, such as raiseIO#, that can raise a (Haskell) synchronous exception to "have_side_effects" but not "can_fail". We must be careful about not discarding such things; see the paper "A semantics for imprecise exceptions". * NB3: *Read* effects (like reading an IORef) don't count here, because it doesn't matter if we don't do them, or do them more than once. *Sequencing* is maintained by the data dependency of the state token. ---------- can_fail ---------------------------- A primop "can_fail" if it can fail with an *unchecked* exception on some elements of its input domain. Main examples: division (fails on zero demoninator) array indexing (fails if the index is out of bounds) An "unchecked exception" is one that is an outright error, (not turned into a Haskell exception,) such as seg-fault or divide-by-zero error. Such can_fail primops are ALWAYS surrounded with a test that checks for the bad cases, but we need to be very careful about code motion that might move it out of the scope of the test. Note [Transformations affected by can_fail and has_side_effects] ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ The can_fail and has_side_effects properties have the following effect on program transformations. Summary table is followed by details. can_fail has_side_effects Discard NO NO Float in YES YES Float out NO NO Duplicate YES NO * Discarding. case (a `op` b) of _ -> rhs ===> rhs You should not discard a has_side_effects primop; e.g. case (writeIntArray# a i v s of (# _, _ #) -> True Arguably you should be able to discard this, since the returned stat token is not used, but that relies on NEVER inlining unsafePerformIO, and programmers sometimes write this kind of stuff by hand (Trac #9390). So we (conservatively) never discard a has_side_effects primop. However, it's fine to discard a can_fail primop. For example case (indexIntArray# a i) of _ -> True We can discard indexIntArray#; it has can_fail, but not has_side_effects; see Trac #5658 which was all about this. Notice that indexIntArray# is (in a more general handling of effects) read effect, but we don't care about that here, and treat read effects as *not* has_side_effects. Similarly (a `/#` b) can be discarded. It can seg-fault or cause a hardware exception, but not a synchronous Haskell exception. Synchronous Haskell exceptions, e.g. from raiseIO#, are treated as has_side_effects and hence are not discarded. * Float in. You can float a can_fail or has_side_effects primop *inwards*, but not inside a lambda (see Duplication below). * Float out. You must not float a can_fail primop *outwards* lest you escape the dynamic scope of the test. Example: case d ># 0# of True -> case x /# d of r -> r +# 1 False -> 0 Here we must not float the case outwards to give case x/# d of r -> case d ># 0# of True -> r +# 1 False -> 0 Nor can you float out a has_side_effects primop. For example: if blah then case writeMutVar# v True s0 of (# s1 #) -> s1 else s0 Notice that s0 is mentioned in both branches of the 'if', but only one of these two will actually be consumed. But if we float out to case writeMutVar# v True s0 of (# s1 #) -> if blah then s1 else s0 the writeMutVar will be performed in both branches, which is utterly wrong. * Duplication. You cannot duplicate a has_side_effect primop. You might wonder how this can occur given the state token threading, but just look at Control.Monad.ST.Lazy.Imp.strictToLazy! We get something like this p = case readMutVar# s v of (# s', r #) -> (S# s', r) s' = case p of (s', r) -> s' r = case p of (s', r) -> r (All these bindings are boxed.) If we inline p at its two call sites, we get a catastrophe: because the read is performed once when s' is demanded, and once when 'r' is demanded, which may be much later. Utterly wrong. Trac #3207 is real example of this happening. However, it's fine to duplicate a can_fail primop. That is really the only difference between can_fail and has_side_effects. Note [Implementation: how can_fail/has_side_effects affect transformations] ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ How do we ensure that that floating/duplication/discarding are done right in the simplifier? Two main predicates on primpops test these flags: primOpOkForSideEffects <=> not has_side_effects primOpOkForSpeculation <=> not (has_side_effects || can_fail) * The "no-float-out" thing is achieved by ensuring that we never let-bind a can_fail or has_side_effects primop. The RHS of a let-binding (which can float in and out freely) satisfies exprOkForSpeculation; this is the let/app invariant. And exprOkForSpeculation is false of can_fail and has_side_effects. * So can_fail and has_side_effects primops will appear only as the scrutinees of cases, and that's why the FloatIn pass is capable of floating case bindings inwards. * The no-duplicate thing is done via primOpIsCheap, by making has_side_effects things (very very very) not-cheap! -} primOpHasSideEffects :: PrimOp -> Bool #include "primop-has-side-effects.hs-incl" primOpCanFail :: PrimOp -> Bool #include "primop-can-fail.hs-incl" primOpOkForSpeculation :: PrimOp -> Bool -- See Note [PrimOp can_fail and has_side_effects] -- See comments with CoreUtils.exprOkForSpeculation -- primOpOkForSpeculation => primOpOkForSideEffects primOpOkForSpeculation op = primOpOkForSideEffects op && not (primOpOutOfLine op || primOpCanFail op) -- I think the "out of line" test is because out of line things can -- be expensive (eg sine, cosine), and so we may not want to speculate them primOpOkForSideEffects :: PrimOp -> Bool primOpOkForSideEffects op = not (primOpHasSideEffects op) {- Note [primOpIsCheap] ~~~~~~~~~~~~~~~~~~~~ @primOpIsCheap@, as used in \tr{SimplUtils.lhs}. For now (HACK WARNING), we just borrow some other predicates for a what-should-be-good-enough test. "Cheap" means willing to call it more than once, and/or push it inside a lambda. The latter could change the behaviour of 'seq' for primops that can fail, so we don't treat them as cheap. -} primOpIsCheap :: PrimOp -> Bool -- See Note [PrimOp can_fail and has_side_effects] primOpIsCheap op = primOpOkForSpeculation op -- In March 2001, we changed this to -- primOpIsCheap op = False -- thereby making *no* primops seem cheap. But this killed eta -- expansion on case (x ==# y) of True -> \s -> ... -- which is bad. In particular a loop like -- doLoop n = loop 0 -- where -- loop i | i == n = return () -- | otherwise = bar i >> loop (i+1) -- allocated a closure every time round because it doesn't eta expand. -- -- The problem that originally gave rise to the change was -- let x = a +# b *# c in x +# x -- were we don't want to inline x. But primopIsCheap doesn't control -- that (it's exprIsDupable that does) so the problem doesn't occur -- even if primOpIsCheap sometimes says 'True'. {- ************************************************************************ * * PrimOp code size * * ************************************************************************ primOpCodeSize ~~~~~~~~~~~~~~ Gives an indication of the code size of a primop, for the purposes of calculating unfolding sizes; see CoreUnfold.sizeExpr. -} primOpCodeSize :: PrimOp -> Int #include "primop-code-size.hs-incl" primOpCodeSizeDefault :: Int primOpCodeSizeDefault = 1 -- CoreUnfold.primOpSize already takes into account primOpOutOfLine -- and adds some further costs for the args in that case. primOpCodeSizeForeignCall :: Int primOpCodeSizeForeignCall = 4 {- ************************************************************************ * * PrimOp types * * ************************************************************************ -} primOpType :: PrimOp -> Type -- you may want to use primOpSig instead primOpType op = case primOpInfo op of Dyadic _occ ty -> dyadic_fun_ty ty Monadic _occ ty -> monadic_fun_ty ty Compare _occ ty -> compare_fun_ty ty GenPrimOp _occ tyvars arg_tys res_ty -> mkForAllTys tyvars (mkFunTys arg_tys res_ty) primOpOcc :: PrimOp -> OccName primOpOcc op = case primOpInfo op of Dyadic occ _ -> occ Monadic occ _ -> occ Compare occ _ -> occ GenPrimOp occ _ _ _ -> occ -- primOpSig is like primOpType but gives the result split apart: -- (type variables, argument types, result type) -- It also gives arity, strictness info primOpSig :: PrimOp -> ([TyVar], [Type], Type, Arity, StrictSig) primOpSig op = (tyvars, arg_tys, res_ty, arity, primOpStrictness op arity) where arity = length arg_tys (tyvars, arg_tys, res_ty) = case (primOpInfo op) of Monadic _occ ty -> ([], [ty], ty ) Dyadic _occ ty -> ([], [ty,ty], ty ) Compare _occ ty -> ([], [ty,ty], intPrimTy) GenPrimOp _occ tyvars arg_tys res_ty -> (tyvars, arg_tys, res_ty ) data PrimOpResultInfo = ReturnsPrim PrimRep | ReturnsAlg TyCon -- Some PrimOps need not return a manifest primitive or algebraic value -- (i.e. they might return a polymorphic value). These PrimOps *must* -- be out of line, or the code generator won't work. getPrimOpResultInfo :: PrimOp -> PrimOpResultInfo getPrimOpResultInfo op = case (primOpInfo op) of Dyadic _ ty -> ReturnsPrim (typePrimRep ty) Monadic _ ty -> ReturnsPrim (typePrimRep ty) Compare _ _ -> ReturnsPrim (tyConPrimRep intPrimTyCon) GenPrimOp _ _ _ ty | isPrimTyCon tc -> ReturnsPrim (tyConPrimRep tc) | otherwise -> ReturnsAlg tc where tc = tyConAppTyCon ty -- All primops return a tycon-app result -- The tycon can be an unboxed tuple, though, which -- gives rise to a ReturnAlg {- We do not currently make use of whether primops are commutable. We used to try to move constants to the right hand side for strength reduction. -} {- commutableOp :: PrimOp -> Bool #include "primop-commutable.hs-incl" -} -- Utils: dyadic_fun_ty, monadic_fun_ty, compare_fun_ty :: Type -> Type dyadic_fun_ty ty = mkFunTys [ty, ty] ty monadic_fun_ty ty = mkFunTy ty ty compare_fun_ty ty = mkFunTys [ty, ty] intPrimTy -- Output stuff: pprPrimOp :: PrimOp -> SDoc pprPrimOp other_op = pprOccName (primOpOcc other_op) {- ************************************************************************ * * \subsubsection[PrimCall]{User-imported primitive calls} * * ************************************************************************ -} data PrimCall = PrimCall CLabelString PackageKey instance Outputable PrimCall where ppr (PrimCall lbl pkgId) = text "__primcall" <+> ppr pkgId <+> ppr lbl