1. 26 Nov, 2008 6 commits
  2. 17 Oct, 2008 1 commit
  3. 16 Oct, 2008 1 commit
    • dias@eecs.harvard.edu's avatar
      Fixed linear regalloc bug, dropped some tracing code · c62b824e
      dias@eecs.harvard.edu authored
      o The linear-scan register allocator sometimes allocated a block
        before allocating one of its predecessors, which could lead
        to inconsistent allocations. Now, we allocate a block only
        if a predecessor has set the "incoming" assignments for the block
        (or if it's the procedure's entry block).
      o Also commented out some tracing code on the new codegen path.
      c62b824e
  4. 14 Oct, 2008 4 commits
    • dias@eecs.harvard.edu's avatar
      Keep update frames live even in functions that never return · 41f7ea2f
      dias@eecs.harvard.edu authored
      An unusual case, but without it:
      (a) we had an assertion failure
      (b) we can overwrite the caller's infotable, which might cause
          the garbage collector to collect live data.
      Better to keep the update frame live at all call sites,
      not just at returns.
      41f7ea2f
    • dias@eecs.harvard.edu's avatar
      Removed space and time inefficiency in procpoint splitting · e13a12b7
      dias@eecs.harvard.edu authored
      I was adding extra jumps to every procpoint, even when the split-off graph
      referred to only some of the procpoints. No effect on correctness,
      but a big effect on space/time efficiency when there are lots of procpoints...
      e13a12b7
    • dias@eecs.harvard.edu's avatar
      Clarify the SRT building process · e367ebeb
      dias@eecs.harvard.edu authored
      Before: building a closure that would build an SRT given the top-level
      SRT. It was somewhat difficult to understand the control flow, and it
      may have had held onto some data structures long after they should be dead.
      Now, I just bundle the info we need about CAFs along with the procedure
      and directly call a new top-level function to build the SRTs later.
      e367ebeb
    • dias@eecs.harvard.edu's avatar
      Don't adjust hp up when the case scrutinee won't allocate · 309f64a0
      dias@eecs.harvard.edu authored
      If the case scrutinee can't allocate, we don't need to do a heap
      check in the case alternatives. (A previous patch got that right.)
      In that case, we had better not adjust the heap pointer to recover
      unused stack space before evaluating the scrutinee -- because we
      aren't going to reallocate for the case alternative.
      309f64a0
  5. 13 Oct, 2008 3 commits
    • dias@eecs.harvard.edu's avatar
      93d047a1
    • dias@eecs.harvard.edu's avatar
      forgot a few files · dc6a72b9
      dias@eecs.harvard.edu authored
      dc6a72b9
    • dias@eecs.harvard.edu's avatar
      Big collection of patches for the new codegen branch. · e6243a81
      dias@eecs.harvard.edu authored
      o Fixed bug that emitted the copy-in code for closure entry
        in the wrong place -- at the initialization of the closure.
      o Refactored some of the closure entry code.
      o Added code to check that no LocalRegs are live-in to a procedure
         -- trip up some buggy programs earlier
      o Fixed environment bindings for thunks
         -- we weren't (re)binding the free variables in a thunk
      o Fixed a bug in proc-point splitting that dropped some updates
        to the entry block in a procedure.
      o Fixed improper calls to code that generates CmmLit's for strings
      o New invariant on cg_loc in CgIdInfo: the expression is always tagged
      o Code to load free vars on entry to a thunk was (wrongly) placed before
        the heap check.
      o Some of the StgCmm code was redundantly passing around Id's
        along with CgIdInfo's; no more.
      o Initialize the LocalReg's that point to a closure before allocating and
        initializing the closure itself -- otherwise, we have problems with
        recursive closure bindings
      o BlockEnv and BlockSet types are now abstract.
      o Update frames:
        - push arguments in Old call area
        - keep track of the return sp in the FCode monad
        - keep the return sp in every call, tail call, and return
            (because it might be different at different call sites,
             e.g. tail calls to the gc after a heap check are performed
                  before pushing the update frame)
        - set the sp appropriately on returns and tail calls
      o Reduce call, tail call, and return to a single LastCall node
      o Added slow entry code, using different calling conventions on entry and tail call
      o More fixes to the calling convention code.
        The tricky stuff is all about the closure environment: it must be passed in R1,
        but in non-closures, there is no such argument, so we can't treat all arguments
        the same way: the closure environment is special. Maybe the right step forward
        would be to define a different calling convention for closure arguments.
      o Let-no-escapes need to be emitted out-of-line -- otherwise, we drop code.
      o Respect RTS requirement of word alignment for pointers
        My stack allocation can pack sub-word values into a single word on the stack,
        but it wasn't requiring word-alignment for pointers. It does now,
        by word-aligning both pointer registers and call areas.
      o CmmLint was over-aggresively ruling out non-word-aligned memory references,
        which may be kosher now that we can spill small values into a single word.
      o Wrong label order on a conditional branch when compiling switches.
      o void args weren't dropped in many cases.
        To help prevent this kind of mistake, I defined a NonVoid wrapper,
        which I'm applying only to Id's for now, although there are probably
        other good candidates.
      o A little code refactoring: separate modules for procpoint analysis splitting, 
        stack layout, and building infotables.
      o Stack limit check: insert along with the heap limit check, using a symbolic
        constant (a special CmmLit), then replace it when the stack layout is known.
      o Removed last node: MidAddToContext 
      o Adding block id as a literal: means that the lowering of the calling conventions
        no longer has to produce labels early, which was inhibiting common-block elimination.
        Will also make it easier for the non-procpoint-splitting path.
      o Info tables: don't try to describe the update frame!
      o Over aggressive use of NonVoid!!!!
        Don't drop the non-void args before setting the type of the closure!!!
      o Sanity checking:
        Added a pass to stub dead dead slots on the stack
        (only ~10 lines with the dataflow framework)
      o More sanity checking:
        Check that incoming pointer arguments are non-stubbed.
        Note: these checks are still subject to dead-code removal, but they should
        still be quite helpful.
      o Better sanity checking: why stop at function arguments?
        Instead, in mkAssign, check that _any_ assignment to a pointer type is non-null
        -- the sooner the crash, the easier it is to debug.
        Still need to add the debugging flag to turn these checks on explicitly.
      o Fixed yet another calling convention bug.
        This time, the calls to the GC were wrong. I've added a new convention
        for GC calls and invoked it where appropriate.
        We should really straighten out the calling convention stuff:
          some of the code (and documentation) is spread across the compiler,
          and there's some magical use of the node register that should really
          be handled (not avoided) by calling conventions.
      o Switch bug: the arms in mkCmmLitSwitch weren't returning to a single join point.
      o Environment shadowing problem in Stg->Cmm:
        When a closure f is bound at the top-level, we should not bind f to the
        node register on entry to the closure.
        Why? Because if the body of f contains a let-bound closure g that refers
        to f, we want to make sure that it refers to the static closure for f.
        Normally, this would all be fine, because when we compile a closure,
        we rebind free variables in the environment. But f doesn't look like
        a free variable because it's a static value. So, the binding for f
        remains in the environment when we compile g, inconveniently referring
        to the wrong thing.
        Now, I bind the variable in the local environment only if the closure is not
        bound at the top level. It's still okay to make assumptions about the
        node holding the closure environment; we just won't find the binding
        in the environment, so code that names the closure will now directly
        get the label of the static closure, not the node register holding a
        pointer to the static closure.
      o Don't generate bogus Cmm code containing SRTs during the STG -> Cmm pass!
        The tables made reference to some labels that don't exist when we compute and
        generate the tables in the back end.
      o Safe foreign calls need some special treatment (at least until we have the integrated
        codegen). In particular:
        o they need info tables
        o they are not procpoints -- the successor had better be in the same procedure
        o we cannot (yet) implement the calling conventions early, which means we have
          to carry the calling-conv info all the way to the end
      o We weren't following the old convention when registering a module.
        Now, we use update frames to push any new modules that have to be registered
        and enter the youngest one on the stack.
        We also use the update frame machinery to specify that the return should pop
        the return address off the stack.
      o At each safe foreign call, an infotable must be at the bottom of the stack,
        and the TSO->sp must point to it.
      o More problems with void args in a direct call to a function:
        We were checking the args (minus voids) to check whether the call was saturated,
        which caused problems when the function really wasn't saturated because it
        took an extra void argument.
      o Forgot to distinguish integer != from floating != during Stg->Cmm
      o Updating slotEnv and areaMap to include safe foreign calls
        The dataflow analyses that produce the slotEnv and areaMap give
        results for each basic block, but we also need the results for
        a safe foreign call, which is a middle node.
        After running the dataflow analysis, we have another pass that
        updates the results to includ any safe foreign calls.
      o Added a static flag for the debugging technique that inserts
        instructions to stub dead slots on the stack and crashes when
        a stubbed value is loaded into a pointer-typed LocalReg.
      o C back end expects to see return continuations before their call sites.
        Sorted the flowgraphs appropriately after splitting.
      o PrimOp calling conventions are special -- unlimited registers, no stack
        Yet another calling convention...
      o More void value problems: if the RHS of a case arm is a void-typed variable,
        don't try to return it.
      o When calling some primOp, they may allocate memory; if so, we need to
        do a heap check when we return from the call.
      e6243a81
  6. 14 Aug, 2008 1 commit
    • dias@eecs.harvard.edu's avatar
      Merging in the new codegen branch · 176fa33f
      dias@eecs.harvard.edu authored
      This merge does not turn on the new codegen (which only compiles
      a select few programs at this point),
      but it does introduce some changes to the old code generator.
      
      The high bits:
      1. The Rep Swamp patch is finally here.
         The highlight is that the representation of types at the
         machine level has changed.
         Consequently, this patch contains updates across several back ends.
      2. The new Stg -> Cmm path is here, although it appears to have a
         fair number of bugs lurking.
      3. Many improvements along the CmmCPSZ path, including:
         o stack layout
         o some code for infotables, half of which is right and half wrong
         o proc-point splitting
      176fa33f
  7. 25 Nov, 2008 3 commits
    • Thomas Schilling's avatar
      Major clean-up of HscMain. · e06951a7
      Thomas Schilling authored
      This patch entails a major restructuring of HscMain and a small bugfix
      to MkIface (which required the restructuring in HscMain).
      
      In MkIface:
      
        - mkIface* no longer outputs orphan warnings directly and also no
          longer quits GHC when -Werror is set.  Instead, errors are
          reported using the common IO (Messages, Maybe result) scheme.
      
      In HscMain:
      
        - Get rid of the 'Comp' monad.  This monad was mostly GhcMonad + two
          reader arguments, a ModSummary for the currently compiled module
          and a possible old interface.  The latter actually lead to a small
          space-leak since only its hash was needed (to check whether the
          newly-generated interface file was the same as the original one).
      
          Functions originally of type 'Comp' now only take the arguments
          that they actually need.  This leads to slighly longer argument
          lists in some places, however, it is now much easier to see what
          is actually going on.
      
        - Get rid of 'myParseModule'.  Rename 'parseFile' to 'hscParse'.
      
        - Join 'deSugarModule' and 'hscDesugar' (keeping the latter).
      
        - Rename 'typecheck{Rename}Module{'}' to 'hscTypecheck{Rename}'.
          One variant keeps the renamed syntax, the other doesn't.
      
        - Parameterise 'HscStatus', so that 'InteractiveStatus' is just a
          different parameterisation of 'HscStatus'.
      
        - 'hscCompile{OneShot,Batch,Nothing,Interactive}' are now
          implemented using a (local) typeclass called 'HsCompiler'.  The
          idea is to make the common structure more obvious.  Using this
          typeclass we now have two functions 'genericHscCompile' (original
          'hscCompiler') and 'genericHscRecompile' (original 'genComp')
          describing the default pipeline.  The methods of the typeclass
          describe a sort of "hook" interface (in OO-terms this would be
          called the "template method" pattern).
      
          One problem with this approach is that we parameterise over the
          /result/ type which, in fact, is not actually different for
          "nothing" and "batch" mode.  To avoid functional dependencies or
          associated types, we use type tags to make them artificially
          different and parameterise the type class over the result type.
          A perhaps better approach might be to use records instead.
          
        - Drop some redundant 'HscEnv' arguments.  These were likely
          different from what 'getSession' would return because during
          compilation we temporarily set the module's DynFlags as well as a
          few other fields.  We now use the 'withTempSession' combinator to
          temporarily change the 'HscEnv' and automatically restore the
          original session after the enclosed action has returned (even in
          case of exceptions).
      
        - Rename 'hscCompile' to 'hscGenHardCode' (since that is what it
          does).
      
      Calls in 'GHC' and 'DriverPipeline' accordingly needed small
      adaptions.
      e06951a7
    • simonpj@microsoft.com's avatar
      Fix Trac #2799: TcType.isOverloadedTy · 52420758
      simonpj@microsoft.com authored
      A missing case (for equality predicates) in isOverloadedTy make
      bindInstsOfLocalFuns/Pats do the wrong thing.  Core Lint nailed it.
      
      Merge to 6.10 branch.
      52420758
    • Simon Marlow's avatar
      Fix #2740: we were missing the free variables on some expressions · cf1ad64a
      Simon Marlow authored
      Particularly boolean expresions: the conditional of an 'if', and
      guards, were missing their free variables.
      cf1ad64a
  8. 21 Nov, 2008 1 commit
  9. 23 Nov, 2008 1 commit
  10. 22 Nov, 2008 4 commits
  11. 21 Nov, 2008 6 commits
    • Thomas Schilling's avatar
      34837046
    • Thomas Schilling's avatar
      eb0e20bd
    • Simon Marlow's avatar
      Use mutator threads to do GC, instead of having a separate pool of GC threads · 3ebcd3de
      Simon Marlow authored
      Previously, the GC had its own pool of threads to use as workers when
      doing parallel GC.  There was a "leader", which was the mutator thread
      that initiated the GC, and the other threads were taken from the pool.
      
      This was simple and worked fine for sequential programs, where we did
      most of the benchmarking for the parallel GC, but falls down for
      parallel programs.  When we have N mutator threads and N cores, at GC
      time we would have to stop N-1 mutator threads and start up N-1 GC
      threads, and hope that the OS schedules them all onto separate cores.
      It practice it doesn't, as you might expect.
      
      Now we use the mutator threads to do GC.  This works quite nicely,
      particularly for parallel programs, where each mutator thread scans
      its own spark pool, which is probably in its cache anyway.
      
      There are some flag changes:
      
        -g<n> is removed (-g1 is still accepted for backwards compat).
        There's no way to have a different number of GC threads than mutator
        threads now.
      
        -q1       Use one OS thread for GC (turns off parallel GC)
        -qg<n>    Use parallel GC for generations >= <n> (default: 1)
      
      Using parallel GC only for generations >=1 works well for sequential
      programs.  Compiling an ordinary sequential program with -threaded and
      running it with -N2 or more should help if you do a lot of GC.  I've
      found that adding -qg0 (do parallel GC for generation 0 too) speeds up
      some parallel programs, but slows down some sequential programs.
      Being conservative, I left the threshold at 1.
      
      ToDo: document the new options.
      3ebcd3de
    • Simon Marlow's avatar
      c373ebdb
    • Thomas Schilling's avatar
      Throw SourceErrors instead of ProgramErrors in main/HeaderInfo. · f7fd7fce
      Thomas Schilling authored
      Parse errors during dependency analysis or options parsing really
      shouldn't kill GHC; this is particularly annoying for GHC API clients.
      f7fd7fce
    • Simon Marlow's avatar
      fix the build when !USE_MMAP · 51e6b90f
      Simon Marlow authored
      51e6b90f
  12. 20 Nov, 2008 4 commits
  13. 19 Nov, 2008 5 commits