{-# LANGUAGE GADTs, DisambiguateRecordFields, BangPatterns #-} module CmmProcPoint ( ProcPointSet, Status(..) , callProcPoints, minimalProcPointSet , splitAtProcPoints, procPointAnalysis , attachContInfoTables ) where import GhcPrelude hiding (last, unzip, succ, zip) import DynFlags import BlockId import CLabel import Cmm import PprCmm () import CmmUtils import CmmInfo import CmmLive import CmmSwitch import Data.List (sortBy) import Maybes import Control.Monad import Outputable import Platform import UniqSupply import Hoopl.Block import Hoopl.Collections import Hoopl.Dataflow import Hoopl.Graph import Hoopl.Label -- Compute a minimal set of proc points for a control-flow graph. -- Determine a protocol for each proc point (which live variables will -- be passed as arguments and which will be on the stack). {- A proc point is a basic block that, after CPS transformation, will start a new function. The entry block of the original function is a proc point, as is the continuation of each function call. A third kind of proc point arises if we want to avoid copying code. Suppose we have code like the following: f() { if (...) { ..1..; call foo(); ..2..} else { ..3..; call bar(); ..4..} x = y + z; return x; } The statement 'x = y + z' can be reached from two different proc points: the continuations of foo() and bar(). We would prefer not to put a copy in each continuation; instead we would like 'x = y + z' to be the start of a new procedure to which the continuations can jump: f_cps () { if (...) { ..1..; push k_foo; jump foo_cps(); } else { ..3..; push k_bar; jump bar_cps(); } } k_foo() { ..2..; jump k_join(y, z); } k_bar() { ..4..; jump k_join(y, z); } k_join(y, z) { x = y + z; return x; } You might think then that a criterion to make a node a proc point is that it is directly reached by two distinct proc points. (Note [Direct reachability].) But this criterion is a bit too simple; for example, 'return x' is also reached by two proc points, yet there is no point in pulling it out of k_join. A good criterion would be to say that a node should be made a proc point if it is reached by a set of proc points that is different than its immediate dominator. NR believes this criterion can be shown to produce a minimum set of proc points, and given a dominator tree, the proc points can be chosen in time linear in the number of blocks. Lacking a dominator analysis, however, we turn instead to an iterative solution, starting with no proc points and adding them according to these rules: 1. The entry block is a proc point. 2. The continuation of a call is a proc point. 3. A node is a proc point if it is directly reached by more proc points than one of its predecessors. Because we don't understand the problem very well, we apply rule 3 at most once per iteration, then recompute the reachability information. (See Note [No simple dataflow].) The choice of the new proc point is arbitrary, and I don't know if the choice affects the final solution, so I don't know if the number of proc points chosen is the minimum---but the set will be minimal. Note [Proc-point analysis] ~~~~~~~~~~~~~~~~~~~~~~~~~~ Given a specified set of proc-points (a set of block-ids), "proc-point analysis" figures out, for every block, which proc-point it belongs to. All the blocks belonging to proc-point P will constitute a single top-level C procedure. A non-proc-point block B "belongs to" a proc-point P iff B is reachable from P without going through another proc-point. Invariant: a block B should belong to at most one proc-point; if it belongs to two, that's a bug. Note [Non-existing proc-points] ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ On some architectures it might happen that the list of proc-points computed before stack layout pass will be invalidated by the stack layout. This will happen if stack layout removes from the graph blocks that were determined to be proc-points. Later on in the pipeline we use list of proc-points to perform [Proc-point analysis], but if a proc-point does not exist anymore then we will get compiler panic. See #8205. -} type ProcPointSet = LabelSet data Status = ReachedBy ProcPointSet -- set of proc points that directly reach the block | ProcPoint -- this block is itself a proc point instance Outputable Status where ppr (ReachedBy ps) | setNull ps = text "" | otherwise = text "reached by" <+> (hsep $ punctuate comma $ map ppr $ setElems ps) ppr ProcPoint = text "" -------------------------------------------------- -- Proc point analysis -- Once you know what the proc-points are, figure out -- what proc-points each block is reachable from -- See Note [Proc-point analysis] procPointAnalysis :: ProcPointSet -> CmmGraph -> LabelMap Status procPointAnalysis procPoints cmmGraph@(CmmGraph {g_graph = graph}) = analyzeCmmFwd procPointLattice procPointTransfer cmmGraph initProcPoints where initProcPoints = mkFactBase procPointLattice [ (id, ProcPoint) | id <- setElems procPoints -- See Note [Non-existing proc-points] , id `setMember` labelsInGraph ] labelsInGraph = labelsDefined graph procPointTransfer :: TransferFun Status procPointTransfer block facts = let label = entryLabel block !fact = case getFact procPointLattice label facts of ProcPoint -> ReachedBy $! setSingleton label f -> f result = map (\id -> (id, fact)) (successors block) in mkFactBase procPointLattice result procPointLattice :: DataflowLattice Status procPointLattice = DataflowLattice unreached add_to where unreached = ReachedBy setEmpty add_to (OldFact ProcPoint) _ = NotChanged ProcPoint add_to _ (NewFact ProcPoint) = Changed ProcPoint -- because of previous case add_to (OldFact (ReachedBy p)) (NewFact (ReachedBy p')) | setSize union > setSize p = Changed (ReachedBy union) | otherwise = NotChanged (ReachedBy p) where union = setUnion p' p ---------------------------------------------------------------------- -- It is worth distinguishing two sets of proc points: those that are -- induced by calls in the original graph and those that are -- introduced because they're reachable from multiple proc points. -- -- Extract the set of Continuation BlockIds, see Note [Continuation BlockIds]. callProcPoints :: CmmGraph -> ProcPointSet callProcPoints g = foldlGraphBlocks add (setSingleton (g_entry g)) g where add :: LabelSet -> CmmBlock -> LabelSet add set b = case lastNode b of CmmCall {cml_cont = Just k} -> setInsert k set CmmForeignCall {succ=k} -> setInsert k set _ -> set minimalProcPointSet :: Platform -> ProcPointSet -> CmmGraph -> UniqSM ProcPointSet -- Given the set of successors of calls (which must be proc-points) -- figure out the minimal set of necessary proc-points minimalProcPointSet platform callProcPoints g = extendPPSet platform g (revPostorder g) callProcPoints extendPPSet :: Platform -> CmmGraph -> [CmmBlock] -> ProcPointSet -> UniqSM ProcPointSet extendPPSet platform g blocks procPoints = let env = procPointAnalysis procPoints g add pps block = let id = entryLabel block in case mapLookup id env of Just ProcPoint -> setInsert id pps _ -> pps procPoints' = foldlGraphBlocks add setEmpty g newPoints = mapMaybe ppSuccessor blocks newPoint = listToMaybe newPoints ppSuccessor b = let nreached id = case mapLookup id env `orElse` pprPanic "no ppt" (ppr id <+> ppr b) of ProcPoint -> 1 ReachedBy ps -> setSize ps block_procpoints = nreached (entryLabel b) -- | Looking for a successor of b that is reached by -- more proc points than b and is not already a proc -- point. If found, it can become a proc point. newId succ_id = not (setMember succ_id procPoints') && nreached succ_id > block_procpoints in listToMaybe $ filter newId $ successors b in case newPoint of Just id -> if setMember id procPoints' then panic "added old proc pt" else extendPPSet platform g blocks (setInsert id procPoints') Nothing -> return procPoints' -- At this point, we have found a set of procpoints, each of which should be -- the entry point of a procedure. -- Now, we create the procedure for each proc point, -- which requires that we: -- 1. build a map from proc points to the blocks reachable from the proc point -- 2. turn each branch to a proc point into a jump -- 3. turn calls and returns into jumps -- 4. build info tables for the procedures -- and update the info table for -- the SRTs in the entry procedure as well. -- Input invariant: A block should only be reachable from a single ProcPoint. -- ToDo: use the _ret naming convention that the old code generator -- used. -- EZY splitAtProcPoints :: DynFlags -> CLabel -> ProcPointSet-> ProcPointSet -> LabelMap Status -> CmmDecl -> UniqSM [CmmDecl] splitAtProcPoints dflags entry_label callPPs procPoints procMap (CmmProc (TopInfo {info_tbls = info_tbls}) top_l _ g@(CmmGraph {g_entry=entry})) = do -- Build a map from procpoints to the blocks they reach let add_block :: LabelMap (LabelMap CmmBlock) -> CmmBlock -> LabelMap (LabelMap CmmBlock) add_block graphEnv b = case mapLookup bid procMap of Just ProcPoint -> add graphEnv bid bid b Just (ReachedBy set) -> case setElems set of [] -> graphEnv [id] -> add graphEnv id bid b _ -> panic "Each block should be reachable from only one ProcPoint" Nothing -> graphEnv where bid = entryLabel b add graphEnv procId bid b = mapInsert procId graph' graphEnv where graph = mapLookup procId graphEnv `orElse` mapEmpty graph' = mapInsert bid b graph let liveness = cmmGlobalLiveness dflags g let ppLiveness pp = filter isArgReg $ regSetToList $ expectJust "ppLiveness" $ mapLookup pp liveness graphEnv <- return $ foldlGraphBlocks add_block mapEmpty g -- Build a map from proc point BlockId to pairs of: -- * Labels for their new procedures -- * Labels for the info tables of their new procedures (only if -- the proc point is a callPP) -- Due to common blockification, we may overestimate the set of procpoints. let add_label map pp = mapInsert pp lbls map where lbls | pp == entry = (entry_label, fmap cit_lbl (mapLookup entry info_tbls)) | otherwise = (block_lbl, guard (setMember pp callPPs) >> Just info_table_lbl) where block_lbl = blockLbl pp info_table_lbl = infoTblLbl pp procLabels :: LabelMap (CLabel, Maybe CLabel) procLabels = foldl' add_label mapEmpty (filter (flip mapMember (toBlockMap g)) (setElems procPoints)) -- In each new graph, add blocks jumping off to the new procedures, -- and replace branches to procpoints with branches to the jump-off blocks let add_jump_block :: (LabelMap Label, [CmmBlock]) -> (Label, CLabel) -> UniqSM (LabelMap Label, [CmmBlock]) add_jump_block (env, bs) (pp, l) = do bid <- liftM mkBlockId getUniqueM let b = blockJoin (CmmEntry bid GlobalScope) emptyBlock jump live = ppLiveness pp jump = CmmCall (CmmLit (CmmLabel l)) Nothing live 0 0 0 return (mapInsert pp bid env, b : bs) add_jumps :: LabelMap CmmGraph -> (Label, LabelMap CmmBlock) -> UniqSM (LabelMap CmmGraph) add_jumps newGraphEnv (ppId, blockEnv) = do let needed_jumps = -- find which procpoints we currently branch to mapFoldr add_if_branch_to_pp [] blockEnv add_if_branch_to_pp :: CmmBlock -> [(BlockId, CLabel)] -> [(BlockId, CLabel)] add_if_branch_to_pp block rst = case lastNode block of CmmBranch id -> add_if_pp id rst CmmCondBranch _ ti fi _ -> add_if_pp ti (add_if_pp fi rst) CmmSwitch _ ids -> foldr add_if_pp rst $ switchTargetsToList ids _ -> rst -- when jumping to a PP that has an info table, if -- tablesNextToCode is off we must jump to the entry -- label instead. jump_label (Just info_lbl) _ | tablesNextToCode dflags = info_lbl | otherwise = toEntryLbl info_lbl jump_label Nothing block_lbl = block_lbl add_if_pp id rst = case mapLookup id procLabels of Just (lbl, mb_info_lbl) -> (id, jump_label mb_info_lbl lbl) : rst Nothing -> rst (jumpEnv, jumpBlocks) <- foldM add_jump_block (mapEmpty, []) needed_jumps -- update the entry block let b = expectJust "block in env" $ mapLookup ppId blockEnv blockEnv' = mapInsert ppId b blockEnv -- replace branches to procpoints with branches to jumps blockEnv'' = toBlockMap $ replaceBranches jumpEnv $ ofBlockMap ppId blockEnv' -- add the jump blocks to the graph blockEnv''' = foldl' (flip addBlock) blockEnv'' jumpBlocks let g' = ofBlockMap ppId blockEnv''' -- pprTrace "g' pre jumps" (ppr g') $ do return (mapInsert ppId g' newGraphEnv) graphEnv <- foldM add_jumps mapEmpty $ mapToList graphEnv let to_proc (bid, g) | bid == entry = CmmProc (TopInfo {info_tbls = info_tbls, stack_info = stack_info}) top_l live g' | otherwise = case expectJust "pp label" $ mapLookup bid procLabels of (lbl, Just info_lbl) -> CmmProc (TopInfo { info_tbls = mapSingleton (g_entry g) (mkEmptyContInfoTable info_lbl) , stack_info=stack_info}) lbl live g' (lbl, Nothing) -> CmmProc (TopInfo {info_tbls = mapEmpty, stack_info=stack_info}) lbl live g' where g' = replacePPIds g live = ppLiveness (g_entry g') stack_info = StackInfo { arg_space = 0 , updfr_space = Nothing , do_layout = True } -- cannot use panic, this is printed by -ddump-cmm -- References to procpoint IDs can now be replaced with the -- infotable's label replacePPIds g = {-# SCC "replacePPIds" #-} mapGraphNodes (id, mapExp repl, mapExp repl) g where repl e@(CmmLit (CmmBlock bid)) = case mapLookup bid procLabels of Just (_, Just info_lbl) -> CmmLit (CmmLabel info_lbl) _ -> e repl e = e -- The C back end expects to see return continuations before the -- call sites. Here, we sort them in reverse order -- it gets -- reversed later. let (_, block_order) = foldl' add_block_num (0::Int, mapEmpty :: LabelMap Int) (revPostorder g) add_block_num (i, map) block = (i + 1, mapInsert (entryLabel block) i map) sort_fn (bid, _) (bid', _) = compare (expectJust "block_order" $ mapLookup bid block_order) (expectJust "block_order" $ mapLookup bid' block_order) procs <- return $ map to_proc $ sortBy sort_fn $ mapToList graphEnv return -- pprTrace "procLabels" (ppr procLabels) -- pprTrace "splitting graphs" (ppr procs) procs splitAtProcPoints _ _ _ _ _ t@(CmmData _ _) = return [t] -- Only called from CmmProcPoint.splitAtProcPoints. NB. does a -- recursive lookup, see comment below. replaceBranches :: LabelMap BlockId -> CmmGraph -> CmmGraph replaceBranches env cmmg = {-# SCC "replaceBranches" #-} ofBlockMap (g_entry cmmg) $ mapMap f $ toBlockMap cmmg where f block = replaceLastNode block $ last (lastNode block) last :: CmmNode O C -> CmmNode O C last (CmmBranch id) = CmmBranch (lookup id) last (CmmCondBranch e ti fi l) = CmmCondBranch e (lookup ti) (lookup fi) l last (CmmSwitch e ids) = CmmSwitch e (mapSwitchTargets lookup ids) last l@(CmmCall {}) = l { cml_cont = Nothing } -- NB. remove the continuation of a CmmCall, since this -- label will now be in a different CmmProc. Not only -- is this tidier, it stops CmmLint from complaining. last l@(CmmForeignCall {}) = l lookup id = fmap lookup (mapLookup id env) `orElse` id -- XXX: this is a recursive lookup, it follows chains -- until the lookup returns Nothing, at which point we -- return the last BlockId -- -------------------------------------------------------------- -- Not splitting proc points: add info tables for continuations attachContInfoTables :: ProcPointSet -> CmmDecl -> CmmDecl attachContInfoTables call_proc_points (CmmProc top_info top_l live g) = CmmProc top_info{info_tbls = info_tbls'} top_l live g where info_tbls' = mapUnion (info_tbls top_info) $ mapFromList [ (l, mkEmptyContInfoTable (infoTblLbl l)) | l <- setElems call_proc_points , l /= g_entry g ] attachContInfoTables _ other_decl = other_decl ---------------------------------------------------------------- {- Note [Direct reachability] Block B is directly reachable from proc point P iff control can flow from P to B without passing through an intervening proc point. -} ---------------------------------------------------------------- {- Note [No simple dataflow] Sadly, it seems impossible to compute the proc points using a single dataflow pass. One might attempt to use this simple lattice: data Location = Unknown | InProc BlockId -- node is in procedure headed by the named proc point | ProcPoint -- node is itself a proc point At a join, a node in two different blocks becomes a proc point. The difficulty is that the change of information during iterative computation may promote a node prematurely. Here's a program that illustrates the difficulty: f () { entry: .... L1: if (...) { ... } else { ... } L2: if (...) { g(); goto L1; } return x + y; } The only proc-point needed (besides the entry) is L1. But in an iterative analysis, consider what happens to L2. On the first pass through, it rises from Unknown to 'InProc entry', but when L1 is promoted to a proc point (because it's the successor of g()), L1's successors will be promoted to 'InProc L1'. The problem hits when the new fact 'InProc L1' flows into L2 which is already bound to 'InProc entry'. The join operation makes it a proc point when in fact it needn't be, because its immediate dominator L1 is already a proc point and there are no other proc points that directly reach L2. -} {- Note [Separate Adams optimization] ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ It may be worthwhile to attempt the Adams optimization by rewriting the graph before the assignment of proc-point protocols. Here are a couple of rules: g() returns to k; g() returns to L; k: CopyIn c ress; goto L: ... ==> ... L: // no CopyIn node here L: CopyIn c ress; And when c == c' and ress == ress', this also: g() returns to k; g() returns to L; k: CopyIn c ress; goto L: ... ==> ... L: CopyIn c' ress' L: CopyIn c' ress' ; In both cases the goal is to eliminate k. -}