CmmProcPoint.hs 20.4 KB
Newer Older
1
{-# LANGUAGE GADTs, DisambiguateRecordFields #-}
2

3 4 5
module CmmProcPoint
    ( ProcPointSet, Status(..)
    , callProcPoints, minimalProcPointSet
Simon Marlow's avatar
Simon Marlow committed
6
    , splitAtProcPoints, procPointAnalysis
7
    , attachContInfoTables
8 9
    )
where
10

11
import Prelude hiding (last, unzip, succ, zip)
12

13
import DynFlags
14 15 16
import BlockId
import CLabel
import Cmm
17
import PprCmm ()
18
import CmmUtils
19
import CmmInfo
20
import CmmLive (cmmGlobalLiveness)
21
import CmmSwitch
22 23 24 25
import Data.List (sortBy)
import Maybes
import Control.Monad
import Outputable
26
import Platform
27 28
import UniqSupply

Simon Marlow's avatar
Simon Marlow committed
29
import Hoopl
30 31 32 33

-- Compute a minimal set of proc points for a control-flow graph.

-- Determine a protocol for each proc point (which live variables will
34
-- be passed as arguments and which will be on the stack).
35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86

{-
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.
87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113



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.
114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129
-}

type ProcPointSet = BlockSet

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 "<not-reached>"
      | otherwise = text "reached by" <+>
                    (hsep $ punctuate comma $ map ppr $ setElems ps)
  ppr ProcPoint = text "<procpt>"

--------------------------------------------------
130
-- Proc point analysis
131

132
procPointAnalysis :: ProcPointSet -> CmmGraph -> UniqSM (BlockEnv Status)
133 134
-- Once you know what the proc-points are, figure out
-- what proc-points each block is reachable from
135 136
-- See Note [Proc-point analysis]
procPointAnalysis procPoints g@(CmmGraph {g_graph = graph}) =
Simon Marlow's avatar
Simon Marlow committed
137
  -- pprTrace "procPointAnalysis" (ppr procPoints) $
138
  return $ dataflowAnalFwdBlocks g initProcPoints lattice forward
139 140 141 142
  where initProcPoints = [(id, ProcPoint) | id <- setElems procPoints,
                                            id `setMember` labelsInGraph ]
                                    -- See Note [Non-existing proc-points]
        labelsInGraph  = labelsDefined graph
143
-- transfer equations
144

145
forward :: FwdTransfer CmmNode Status
Simon Marlow's avatar
Simon Marlow committed
146
forward = mkFTransfer3 first middle last
147
    where
Simon Marlow's avatar
Simon Marlow committed
148
      first :: CmmNode C O -> Status -> Status
Peter Wortmann's avatar
Peter Wortmann committed
149
      first (CmmEntry id _) ProcPoint = ReachedBy $ setSingleton id
Simon Marlow's avatar
Simon Marlow committed
150 151 152 153 154 155
      first  _ x = x

      middle _ x = x

      last :: CmmNode O C -> Status -> FactBase Status
      last l x = mkFactBase lattice $ map (\id -> (id, x)) (successors l)
156

157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175
lattice :: DataflowLattice Status
lattice = DataflowLattice "direct proc-point reachability" unreached add_to
    where unreached = ReachedBy setEmpty
          add_to _ (OldFact ProcPoint) _ = (NoChange, ProcPoint)
          add_to _ _ (NewFact ProcPoint) = (SomeChange, ProcPoint)
                       -- because of previous case
          add_to _ (OldFact (ReachedBy p)) (NewFact (ReachedBy p'))
             | setSize union > setSize p = (SomeChange, ReachedBy union)
             | otherwise                 = (NoChange, 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].
176 177 178 179 180 181 182 183
callProcPoints      :: CmmGraph -> ProcPointSet
callProcPoints g = foldGraphBlocks add (setSingleton (g_entry g)) g
  where add :: CmmBlock -> BlockSet -> BlockSet
        add b set = case lastNode b of
                      CmmCall {cml_cont = Just k} -> setInsert k set
                      CmmForeignCall {succ=k}     -> setInsert k set
                      _ -> set

184
minimalProcPointSet :: Platform -> ProcPointSet -> CmmGraph
185
                    -> UniqSM ProcPointSet
186 187
-- Given the set of successors of calls (which must be proc-points)
-- figure out the minimal set of necessary proc-points
188 189
minimalProcPointSet platform callProcPoints g
  = extendPPSet platform g (postorderDfs g) callProcPoints
190

191
extendPPSet :: Platform -> CmmGraph -> [CmmBlock] -> ProcPointSet -> UniqSM ProcPointSet
192
extendPPSet platform g blocks procPoints =
193
    do env <- procPointAnalysis procPoints g
Simon Marlow's avatar
Simon Marlow committed
194
       -- pprTrace "extensPPSet" (ppr env) $ return ()
195 196 197 198 199 200 201 202 203
       let add block pps = let id = entryLabel block
                           in  case mapLookup id env of
                                 Just ProcPoint -> setInsert id pps
                                 _ -> pps
           procPoints' = foldGraphBlocks add setEmpty g
           newPoints = mapMaybe ppSuccessor blocks
           newPoint  = listToMaybe newPoints
           ppSuccessor b =
               let nreached id = case mapLookup id env `orElse`
Ian Lynagh's avatar
Ian Lynagh committed
204
                                       pprPanic "no ppt" (ppr id <+> ppr b) of
205 206 207 208 209 210 211 212 213 214 215 216 217 218 219
                                   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
{-
       case newPoints of
           []  -> return procPoints'
           pps -> extendPPSet g blocks
                    (foldl extendBlockSet procPoints' pps)
-}
220 221 222 223 224 225
       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'
226 227 228 229 230 231 232 233 234 235 236 237


-- 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.
238 239
-- ToDo: use the _ret naming convention that the old code generator
-- used. -- EZY
240
splitAtProcPoints :: DynFlags -> CLabel -> ProcPointSet-> ProcPointSet -> BlockEnv Status ->
241
                     CmmDecl -> UniqSM [CmmDecl]
242
splitAtProcPoints dflags entry_label callPPs procPoints procMap
243
                  (CmmProc (TopInfo {info_tbls = info_tbls})
244
                           top_l _ g@(CmmGraph {g_entry=entry})) =
245 246 247 248 249 250 251
  do -- Build a map from procpoints to the blocks they reach
     let addBlock b graphEnv =
           case mapLookup bid procMap of
             Just ProcPoint -> add graphEnv bid bid b
             Just (ReachedBy set) ->
               case setElems set of
                 []   -> graphEnv
252
                 [id] -> add graphEnv id bid b
253
                 _    -> panic "Each block should be reachable from only one ProcPoint"
254
             Nothing -> graphEnv
255 256 257 258
           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
259

260 261 262 263 264
     let liveness = cmmGlobalLiveness dflags g
     let ppLiveness pp = filter isArgReg $
                         regSetToList $
                         expectJust "ppLiveness" $ mapLookup pp liveness

265
     graphEnv <- return $ foldGraphBlocks addBlock emptyBlockMap g
266

267 268
     -- Build a map from proc point BlockId to pairs of:
     --  * Labels for their new procedures
269 270
     --  * Labels for the info tables of their new procedures (only if
     --    the proc point is a callPP)
271
     -- Due to common blockification, we may overestimate the set of procpoints.
272
     let add_label map pp = mapInsert pp lbls map
273 274 275 276
           where lbls | pp == entry = (entry_label, fmap cit_lbl (mapLookup entry info_tbls))
                      | otherwise   = (block_lbl, guard (setMember pp callPPs) >>
                                                    Just (toInfoLbl block_lbl))
                      where block_lbl = blockLbl pp
277 278 279

         procLabels :: LabelMap (CLabel, Maybe CLabel)
         procLabels = foldl add_label mapEmpty
280
                            (filter (flip mapMember (toBlockMap g)) (setElems procPoints))
281

282 283 284 285
     -- 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 (env, bs) (pp, l) =
           do bid <- liftM mkBlockId getUniqueM
Peter Wortmann's avatar
Peter Wortmann committed
286
              let b = blockJoin (CmmEntry bid GlobalScope) emptyBlock jump
287 288
                  live = ppLiveness pp
                  jump = CmmCall (CmmLit (CmmLabel l)) Nothing live 0 0 0
289
              return (mapInsert pp bid env, b : bs)
290 291

         add_jumps newGraphEnv (ppId, blockEnv) =
292 293 294 295 296 297
           do let needed_jumps = -- find which procpoints we currently branch to
                    mapFold 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
298
                      CmmCondBranch _ ti fi _ -> add_if_pp ti (add_if_pp fi rst)
299
                      CmmSwitch _ ids       -> foldr add_if_pp rst $ switchTargetsToList ids
300
                      _                     -> rst
301 302 303 304 305 306 307 308 309 310 311

                  -- 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
312
                                       Nothing                 -> rst
313 314 315 316 317 318 319 320 321
              (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 insertBlock) blockEnv'' jumpBlocks
322
              let g' = ofBlockMap ppId blockEnv'''
323 324
              -- pprTrace "g' pre jumps" (ppr g') $ do
              return (mapInsert ppId g' newGraphEnv)
325

326
     graphEnv <- foldM add_jumps emptyBlockMap $ mapToList graphEnv
327

328 329 330 331
     let to_proc (bid, g)
             | bid == entry
             =  CmmProc (TopInfo {info_tbls  = info_tbls,
                                  stack_info = stack_info})
332
                        top_l live g'
333 334 335 336 337
             | 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})
338
                               lbl live g'
339 340
                 (lbl, Nothing)
                    -> CmmProc (TopInfo {info_tbls = mapEmpty, stack_info=stack_info})
341
                               lbl live g'
342
                where
343 344
                 g' = replacePPIds g
                 live = ppLiveness (g_entry g')
345 346 347
                 stack_info = StackInfo { arg_space = 0
                                        , updfr_space =  Nothing
                                        , do_layout = True }
348
                               -- cannot use panic, this is printed by -ddump-cmm
349 350 351 352 353

         -- References to procpoint IDs can now be replaced with the
         -- infotable's label
         replacePPIds g = {-# SCC "replacePPIds" #-}
                          mapGraphNodes (id, mapExp repl, mapExp repl) g
354
           where repl e@(CmmLit (CmmBlock bid)) =
355
                   case mapLookup bid procLabels of
356 357
                     Just (_, Just info_lbl)  -> CmmLit (CmmLabel info_lbl)
                     _ -> e
358
                 repl e = e
359 360 361 362

     -- The C back end expects to see return continuations before the
     -- call sites.  Here, we sort them in reverse order -- it gets
     -- reversed later.
363 364 365 366 367 368 369 370 371
     let (_, block_order) = foldl add_block_num (0::Int, emptyBlockMap) (postorderDfs 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
372
splitAtProcPoints _ _ _ _ _ t@(CmmData _ _) = return [t]
373

374 375 376
-- Only called from CmmProcPoint.splitAtProcPoints. NB. does a
-- recursive lookup, see comment below.
replaceBranches :: BlockEnv BlockId -> CmmGraph -> CmmGraph
377 378 379
replaceBranches env cmmg
  = {-# SCC "replaceBranches" #-}
    ofBlockMap (g_entry cmmg) $ mapMap f $ toBlockMap cmmg
380
  where
381 382
    f block = replaceLastNode block $ last (lastNode block)

383 384
    last :: CmmNode O C -> CmmNode O C
    last (CmmBranch id)          = CmmBranch (lookup id)
385
    last (CmmCondBranch e ti fi l) = CmmCondBranch e (lookup ti) (lookup fi) l
386
    last (CmmSwitch e ids)       = CmmSwitch e (mapSwitchTargets lookup ids)
387 388 389 390
    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.
391 392 393 394 395 396
    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

397 398 399 400
-- --------------------------------------------------------------
-- Not splitting proc points: add info tables for continuations

attachContInfoTables :: ProcPointSet -> CmmDecl -> CmmDecl
401 402
attachContInfoTables call_proc_points (CmmProc top_info top_l live g)
 = CmmProc top_info{info_tbls = info_tbls'} top_l live g
403 404 405 406 407 408 409 410
 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

411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429
----------------------------------------------------------------

{-
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
430
                | ProcPoint      -- node is itself a proc point
431

432
At a join, a node in two different blocks becomes a proc point.
433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466
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:

467 468 469 470 471 472
  g() returns to k;                    g() returns to L;
  k: CopyIn c ress; goto L:
   ...                        ==>        ...
  L: // no CopyIn node here            L: CopyIn c ress;


473 474
And when c == c' and ress == ress', this also:

475 476 477 478
  g() returns to k;                    g() returns to L;
  k: CopyIn c ress; goto L:
   ...                        ==>        ...
  L: CopyIn c' ress'                   L: CopyIn c' ress' ;
479 480 481

In both cases the goal is to eliminate k.
-}