Update of ooc-sources available

[Juergen Zimmermann <jnzimmer@informatik.uni-kl.de>]

Hello,
the latest copy of the ooc-Source tree is available at the
well known location
ftp://cognac.informatik.uni-kl.de/pub/ooc/


Here's the relevant part of the ChangeLog:


Wed Jan 24 16:04:20 1996  Juergen Zimmermann  <jnzimmer@informatik.uni-kl.de>

        * optimizers/AlgebraicTransformation.Mod:
        Added the handling of some more cases for algebraic transformations,
        mainly in the SET area.

        * optimizers/DependencyAnalysis.Mod: Changed some comments...

        * optimizers/ConstPropagation.Mod:
        Fixed serious bug which caused regions marked as non-reachable to
        be considered as a valid path for `merge'.

Wed Jan 24 12:56:00 1996  Michael van Acken  <acken@informatik.uni-kl.de>

        * frontend/ParseExpr.Mod (Factor):
        Set current value of `x' to some value, even if no valid
        factor can be found.  This avoids NIL deref later on.

Wed Jan 24 11:56:21 1996  Juergen Zimmermann  <jnzimmer@informatik.uni-kl.de>

        * optimizers/ConstPropagation.Mod: Cleaned up code.
        Reworked all stuff which handles `merges'. Now should not remove
        needed merges, but should catch all unneeded merges.

Fri Jan 19 19:51:29 1996  Frank Copeland  <fjc@wossname.apana.org.au>

        * backend/m68k/OOCRev.Mod: Initial revision

        * backend/m68k/lib/Rts.Mod:
        - [bug] Crashed with a NIL pointer trap if there were no
          parameters on the command line.

        * backend/m68k/lib/THISDIR: Initial revision

        * backend/m68k/OOC.Mod: - Removed log parameter.
        - Changed banner to display revision.

        * backend/m68k/StdTypes.Mod: - Removed logging code.
        - Exported type size constants.

Fri Jan 19 16:22:46 1996  Michael van Acken  <acken@informatik.uni-kl.de>

        * frontend/ParseDecl.Mod:
        `StructModifiers' is called before `StdTypes.StructAlloc'
        (FormalPars): Added calls to `StructModifiers'.
        (ProcDecl): Moved call to `ObjectModifiers' (syntactically) before the
        procedure identifier.

        * frontend/doc/Modifiers:
        Added text as motivation for `modifiers'.  Split procedure modifiers
        in a part associated with the proc, and a part associated with its
        formal parameters.

        * frontend/GateReduction.Mod (TraverseGates):
        Didn't take undefined values into account.  These
        values are of the type D.Object, instead of D.Const or D.Result.

        * frontend/Data.Mod (InitInstruction): Field `flags' wasn't cleared.

Thu Jan 18 17:59:46 1996  Michael van Acken  <acken@informatik.uni-kl.de>

        * frontend/ParseStatm.Mod (RemoveCollectOpnds):
        Don't remove the first two operands from the
        collect instruction.  They hold the values of $mem and $store.

Thu Jan 18 09:54:36 1996  Michael van Acken  <acken@informatik.uni-kl.de>

        * frontend/ParseExpr.Mod, frontend/ParseStatm.Mod:
        Rewrote code generation for boolean expressions.  The idea is to model
        the data flow through conditional expressions (& resp. OR) as precise as
        possible.  As a side-effect the handling of negated expressions
        improved: the laws of De Morgan are automatically applied on negated &
        resp. OR, and negations used in guards are removed by swapping the
        `true' and the `false' path.

Wed Jan 17 15:22:55 1996  Frank Copeland  <fjc@wossname.apana.org.au>

        * backend/m68k/OOC.Mod: - Updated to track additions in OOC-960108.

        * backend/m68k/lib/TimeStamp.Mod, backend/m68k/lib/Strings.Mod, backend/
m68k/lib/RandomNumbers.Mod:
        Initial revision

Wed Jan 17 12:58:21 1996  Juergen Zimmermann  <jnzimmer@informatik.uni-kl.de>

        * optimizers/DependencyAnalysis.Mod: Initial revision

Tue Jan 16 16:25:55 1996  Juergen Zimmermann  <jnzimmer@informatik.uni-kl.de>

        * optimizers/AlgebraicTransformation.Mod:
        The module is still not finished.

        Introduced `ArithmTransform', a procedure for all those special
        arithmetical cases which were not catched yet.

        * optimizers/ConstPropagation.Mod:
        Also have a look on `mergeCond' the same way as we do on `mergeIf'/`merg
eCase'.
        This way, some more `gates' become eliminated.

        Rewrote parts of the code and replaced them by smaller, equivalent ones.