Generate .loc/.file directives from source ticks

This generates DWARF, albeit indirectly using the assembler. This is
the easiest (and, apparently, quite standard) method of generating the
.debug_line DWARF section.

Notes:

* Note we have to make sure that .file directives appear correctly
  before the respective .loc. Right now we ppr them manually, which makes
  them absent from dumps. Fixing this would require .file to become a
  native instruction.

* We have to pass a lot of things around the native code generator. I
  know Ian did quite a bit of refactoring already, but having one common
  monad could *really* simplify things here...

* To support SplitObjcs, we need to emit/reset all DWARF data at every
  split. We use the occassion to move split marker generation to
  cmmNativeGenStream as well, so debug data extraction doesn't have to
  choke on it.

(From Phabricator D396)

1 files changed, 78 insertions, 51 deletions

diff --git a/compiler/nativeGen/AsmCodeGen.hs b/compiler/nativeGen/AsmCodeGen.hs
index 55d1247952..daaeaa217c 100644
--- a/compiler/nativeGen/AsmCodeGen.hs
+++ b/compiler/nativeGen/AsmCodeGen.hs
@@ -266,6 +266,7 @@ data NativeGenAcc statics instr
         , ngs_linearStats :: ![[Linear.RegAllocStats]]
         , ngs_labels      :: ![Label]
         , ngs_debug       :: ![DebugBlock]
+        , ngs_dwarfFiles  :: !DwarfFiles
         }
 
 nativeCodeGen' :: (Outputable statics, Outputable instr, Instruction instr)
@@ -278,25 +279,17 @@ nativeCodeGen' :: (Outputable statics, Outputable instr, Instruction instr)
                -> IO UniqSupply
 nativeCodeGen' dflags this_mod modLoc ncgImpl h us cmms
  = do
-        let split_cmms  = Stream.map add_split cmms
         -- BufHandle is a performance hack.  We could hide it inside
         -- Pretty if it weren't for the fact that we do lots of little
         -- printDocs here (in order to do codegen in constant space).
         bufh <- newBufHandle h
+        let ngs0 = NGS [] [] [] [] [] [] emptyUFM
         (ngs, us') <- cmmNativeGenStream dflags this_mod modLoc ncgImpl bufh us
-                                         split_cmms (NGS [] [] [] [] [] [])
+                                         cmms ngs0
         finishNativeGen dflags bufh ngs
 
         return us'
 
- where  add_split tops
-                | gopt Opt_SplitObjs dflags = split_marker : tops
-                | otherwise                 = tops
-
-        split_marker = CmmProc mapEmpty mkSplitMarkerLabel []
-                               (ofBlockList (panic "split_marker_entry") [])
-
-
 finishNativeGen :: Instruction instr
                 => DynFlags
                 -> BufHandle
@@ -355,56 +348,78 @@ cmmNativeGenStream :: (Outputable statics, Outputable instr, Instruction instr)
 cmmNativeGenStream dflags this_mod modLoc ncgImpl h us cmm_stream ngs
  = do r <- Stream.runStream cmm_stream
       case r of
-          Left () ->
-              return (ngs { ngs_imports = reverse $ ngs_imports ngs
-                          , ngs_natives = reverse $ ngs_natives ngs
-                          , ngs_colorStats = reverse $ ngs_colorStats ngs
-                          , ngs_linearStats = reverse $ ngs_linearStats ngs
-                          },
-                      us)
-          Right (cmms, cmm_stream') -> do
-
-              -- Generate debug information
-              let debugFlag = gopt Opt_Debug dflags
-                  !ndbgs | debugFlag = cmmDebugGen modLoc cmms
-                         | otherwise = []
-
-              -- Generate native code
-              (ngs',us') <- cmmNativeGens dflags this_mod ncgImpl h us cmms ngs 0
-
-              -- Link native code information into debug blocks
-              let !ldbgs = cmmDebugLink (ngs_labels ngs') ndbgs
-              dumpIfSet_dyn dflags Opt_D_dump_debug "Debug Infos"
-                (vcat $ map ppr ldbgs)
-
-              -- Strip references to native code unless we want to dump it later
-              let ngs'' = ngs' { ngs_debug  = ngs_debug ngs' ++ ldbgs
-                               , ngs_labels = [] }
-              cmmNativeGenStream dflags this_mod modLoc ncgImpl h us'
-                  cmm_stream' ngs''
+        Left () ->
+          return (ngs { ngs_imports = reverse $ ngs_imports ngs
+                      , ngs_natives = reverse $ ngs_natives ngs
+                      , ngs_colorStats = reverse $ ngs_colorStats ngs
+                      , ngs_linearStats = reverse $ ngs_linearStats ngs
+                      },
+                  us)
+        Right (cmms, cmm_stream') -> do
+
+          -- Generate debug information
+          let debugFlag = gopt Opt_Debug dflags
+              !ndbgs | debugFlag = cmmDebugGen modLoc cmms
+                     | otherwise = []
+              dbgMap = debugToMap ndbgs
+
+          -- Insert split marker, generate native code
+          let splitFlag = gopt Opt_SplitObjs dflags
+              split_marker = CmmProc mapEmpty mkSplitMarkerLabel [] $
+                             ofBlockList (panic "split_marker_entry") []
+              cmms' | splitFlag  = split_marker : cmms
+                    | otherwise  = cmms
+          (ngs',us') <- cmmNativeGens dflags this_mod modLoc ncgImpl h dbgMap us
+                                      cmms' ngs 0
+
+          -- Link native code information into debug blocks
+          let !ldbgs = cmmDebugLink (ngs_labels ngs') ndbgs
+          dumpIfSet_dyn dflags Opt_D_dump_debug "Debug Infos"
+            (vcat $ map ppr ldbgs)
+
+          -- Clear DWARF info when generating split object files
+          let ngs'' | debugFlag && splitFlag
+                    = ngs' { ngs_debug = []
+                           , ngs_dwarfFiles = emptyUFM
+                           , ngs_labels = [] }
+                    | otherwise
+                    = ngs' { ngs_debug  = ngs_debug ngs' ++ ldbgs
+                           , ngs_labels = [] }
+          cmmNativeGenStream dflags this_mod modLoc ncgImpl h us'
+              cmm_stream' ngs''
 
 -- | Do native code generation on all these cmms.
 --
 cmmNativeGens :: (Outputable statics, Outputable instr, Instruction instr)
               => DynFlags
-              -> Module
+              -> Module -> ModLocation
               -> NcgImpl statics instr jumpDest
               -> BufHandle
+              -> LabelMap DebugBlock
               -> UniqSupply
               -> [RawCmmDecl]
               -> NativeGenAcc statics instr
               -> Int
               -> IO (NativeGenAcc statics instr, UniqSupply)
 
-cmmNativeGens _ _ _ _ us [] ngs !_
+cmmNativeGens _ _ _ _ _ _ us [] ngs !_
         = return (ngs, us)
 
-cmmNativeGens dflags this_mod ncgImpl h us (cmm : cmms) ngs count
+cmmNativeGens dflags this_mod modLoc ncgImpl h dbgMap us
+              (cmm : cmms) ngs count
  = do
-        (us', native, imports, colorStats, linearStats)
-                <- {-# SCC "cmmNativeGen" #-} cmmNativeGen dflags this_mod ncgImpl us cmm count
+        let fileIds = ngs_dwarfFiles ngs
+        (us', fileIds', native, imports, colorStats, linearStats)
+          <- {-# SCC "cmmNativeGen" #-}
+             cmmNativeGen dflags this_mod modLoc ncgImpl us fileIds dbgMap
+                          cmm count
+
+        let newFileIds = fileIds' `minusUFM` fileIds
+            pprDecl (f,n) = ptext (sLit "\t.file ") <> ppr n <+>
+                            doubleQuotes (ftext f)
 
         emitNativeCode dflags h $ vcat $
+          map pprDecl (eltsUFM newFileIds) ++
           map (pprNatCmmDecl ncgImpl) native
 
         -- force evaluation all this stuff to avoid space leaks
@@ -420,8 +435,10 @@ cmmNativeGens dflags this_mod ncgImpl h us (cmm : cmms) ngs count
                       , ngs_colorStats  = colorStats `mCon` ngs_colorStats ngs
                       , ngs_linearStats = linearStats `mCon` ngs_linearStats ngs
                       , ngs_labels      = ngs_labels ngs ++ labels'
+                      , ngs_dwarfFiles  = fileIds'
                       }
-        cmmNativeGens dflags this_mod ncgImpl h us' cmms ngs' (count + 1)
+        cmmNativeGens dflags this_mod modLoc ncgImpl h dbgMap us'
+                      cmms ngs' (count + 1)
 
  where  seqString []            = ()
         seqString (x:xs)        = x `seq` seqString xs
@@ -444,18 +461,21 @@ emitNativeCode dflags h sdoc = do
 cmmNativeGen
         :: (Outputable statics, Outputable instr, Instruction instr)
     => DynFlags
-    -> Module
+    -> Module -> ModLocation
     -> NcgImpl statics instr jumpDest
         -> UniqSupply
+        -> DwarfFiles
+        -> LabelMap DebugBlock
         -> RawCmmDecl                                   -- ^ the cmm to generate code for
         -> Int                                          -- ^ sequence number of this top thing
         -> IO   ( UniqSupply
+                , DwarfFiles
                 , [NatCmmDecl statics instr]                -- native code
                 , [CLabel]                                  -- things imported by this cmm
                 , Maybe [Color.RegAllocStats statics instr] -- stats for the coloring register allocator
                 , Maybe [Linear.RegAllocStats])             -- stats for the linear register allocators
 
-cmmNativeGen dflags this_mod ncgImpl us cmm count
+cmmNativeGen dflags this_mod modLoc ncgImpl us fileIds dbgMap cmm count
  = do
         let platform = targetPlatform dflags
 
@@ -474,9 +494,11 @@ cmmNativeGen dflags this_mod ncgImpl us cmm count
                 (pprCmmGroup [opt_cmm])
 
         -- generate native code from cmm
-        let ((native, lastMinuteImports), usGen) =
+        let ((native, lastMinuteImports, fileIds'), usGen) =
                 {-# SCC "genMachCode" #-}
-                initUs us $ genMachCode dflags this_mod (cmmTopCodeGen ncgImpl) opt_cmm
+                initUs us $ genMachCode dflags this_mod modLoc
+                                        (cmmTopCodeGen ncgImpl)
+                                        fileIds dbgMap opt_cmm
 
         dumpIfSet_dyn dflags
                 Opt_D_dump_asm_native "Native code"
@@ -607,6 +629,7 @@ cmmNativeGen dflags this_mod ncgImpl us cmm count
                 (vcat $ map (pprNatCmmDecl ncgImpl) expanded)
 
         return  ( usAlloc
+                , fileIds'
                 , expanded
                 , lastMinuteImports ++ imports
                 , ppr_raStatsColor
@@ -862,21 +885,25 @@ apply_mapping ncgImpl ufm (CmmProc info lbl live (ListGraph blocks))
 
 genMachCode
         :: DynFlags
-        -> Module
+        -> Module -> ModLocation
         -> (RawCmmDecl -> NatM [NatCmmDecl statics instr])
+        -> DwarfFiles
+        -> LabelMap DebugBlock
         -> RawCmmDecl
         -> UniqSM
                 ( [NatCmmDecl statics instr]
-                , [CLabel])
+                , [CLabel]
+                , DwarfFiles)
 
-genMachCode dflags this_mod cmmTopCodeGen cmm_top
+genMachCode dflags this_mod modLoc cmmTopCodeGen fileIds dbgMap cmm_top
   = do  { initial_us <- getUniqueSupplyM
         ; let initial_st           = mkNatM_State initial_us 0 dflags this_mod
+                                                  modLoc fileIds dbgMap
               (new_tops, final_st) = initNat initial_st (cmmTopCodeGen cmm_top)
               final_delta          = natm_delta final_st
               final_imports        = natm_imports final_st
         ; if   final_delta == 0
-          then return (new_tops, final_imports)
+          then return (new_tops, final_imports, natm_fileid final_st)
           else pprPanic "genMachCode: nonzero final delta" (int final_delta)
     }