Modula-2 and Oberon-2 Compiler for HP OpenVMS Alpha and VAX

The programming languages Modula-2 and Oberon-2

Modula-2

The Modula-2 language reference

Oberon-2

The Oberon language reference

The Oberon-2 language reference

All above mentioned books are also available in German.

• Programming in Oberon (2004) - A derivative of Programming in Modula-2 (1982) - N. Wirth [PDF (334 KB)]
• Algorithms and Data Structures (1985) (Oberon version: August 2004) - N. Wirth [PDF (1'241 KB)]
• Niklaus Wirth: Compiler Construction. Addison-Wesley, 1996. [PDF file (6.2 MB)]

Modula-2 and Oberon-2 Compiler for OpenVMS

These products were developed and are maintained by ModulaWare

Modula-2 and Oberon-2 native-code compiler for HP Alpha and VAX under the OpenVMS operating system

• MaX V5.05 32 bit and 64 bit native code Modula-2 compiler for HP OpenVMS Alpha version 6.2 to 8.3
• MaX V4.18 32 bit native code Modula-2 compiler for HP OpenVMS Alpha
• MVR V4.16 32 bit native code Modula-2 compiler for HP OpenVMS VAX
• A2O V3.00 64 bit native code Oberon-2 compiler for HP OpenVMS Alpha

Common features of MaX/MVR/A2O:

• complies with OpenVMS procedure calling conventions
• generates position independent, reentrant, native-code
• is Y2K-compliant
• supports VAX floating point types and arithmetic: F, D, and G_FLOATING, plus H_FLOATING on VAX; LONGREAL is D_FLOATING (or T_FLOATING on Alpha with /IEEE)
• supports VAX complex types (and complex arithmetic in Modula-2): F, D, and G_COMPLEX, plus H_COMPLEX on VAX; LONGCOMPLEX is D_COMPLEX (or T_COMPLEX on Alpha with /IEEE)
• fully supports OpenVMS multi-process, symbolic run-time debugger, e.g.: step, step/into, watch and break point, examine simple and structured variables (record, array, and pointer-types), screen-mode, etc.
• ISO Modula-2 library modules
• representation of floating point literals and constant expressions of any floating point type (and complex type in Modula-2)
• allows structured function results, i.e.: function return type can be of basic-, array- and record-type
• sets with up-to 128 elements, which allows SET OF CHAR [CHR(0)..CHR(127)]
• packed and aligned record fields (/packed_records)
• uses static link chain technique to access outer-scope local variables in nested procedures (provides for programming of exception handlers)
• detection of tangling pointers (standard in Oberon-2, /optimize=(clearstack) in Modula-2); clears local variables declared in procedures and in dynamic modules in the procedure/module-body entry code. Global (statically allocated) variables are cleared by a OpenVMS linker directive. The heap is cleared by run-time system in Oberon-2 and in Modula-2; the heap can be cleared by a customized ALLOCATE. So any unitialized pointer access will produce an access-violation exception
• allows to call foreign procedures and functions via foreign definition modules
• supports OpenVMS exception handling (ESTABLISH)
• features coroutines for multi-programming (with very fast context-switch and secure stack-overflow check on Alpha)
• comes with VMS, RMS, MTH, RTL, DPML, OTS, LIB, STR, PPL, OSF/Motif, DecWindows/X11 foreign language interface modules [provided as is/unsupported]
• allows programming of asynchronous system traps (AST)
• built-in system function REGISTER in Modula-2 and GETREG/PUTREG in Oberon-2
• built-in cross-reference (/cross_reference)
• built-in disassembler (/machine_code)
• includes make-tool
• supports language sensitive editor (LSE)
• features persistent objects (Oberon-2 only)
• includes source code analyser (Oberon-2 only)
• distribution kit includes source code of many examples and library modules
• simple installation procedure; complete library is compiled at installation time
• additionally on Alpha:
  • 32 bit and 64 bit pointers to access very large memory (VLM) in Oberon-2 and Modula-2 (V5)
    For details about VLM see the virtual region programming example and the
    Digital Technical Journal:
    
    • Extending OpenVMS for 64-bit Addressable Virtual Memory, October 1996, Michael S. Harvey and Leonard S. Szubowicz (Volume 8 Number 2) Pdf
    • The OpenVMS Mixed Pointer Size Environment, October 1996, by Thomas R. Benson, Karen L. Noel and Richard E. Peterson: Pdf
    • Adding 64-bit Pointer Support for a 32-bit Run-time Library, October 1996, by Duane A. Smith Pdf
    • OpenVMS Alpha 64-bit Very Large Memory Design, by Karen L. Noel , and Nitin Y. Karkhanis, 1997 (Volume 9 Number 4) Pdf
  • IEEE floating point types and arithmetic: S, T_FLOATING
  • IEEE complex types (and arithmetic in Modula-2): S, T_COMPLEX
  • selectable default floating point type for [LONG]REAL is either VAX- or IEEE-floating (/[NO]IEEE)
  • default floating point type determines the base type for the complex type [LONG]COMPLEX
  • 32 bit and 64 bit integer type and arithmetic
  • Modula-2 additionally: 64 bit cardinal type and arithmetic
  • Oberon-2 additionally: 8 and 16 bit integer data types and arithmetic
  • Oberon-2 and Modula-2 (V5): 64 bit precision for whole number expressions and constant literals
  • built-in instruction scheduler
• ISO Modula-2 language extensions (VAX and Alpha):
  • additional "/" and "REM" integer and cardinal operators
  • complex arithmetic, relational operators and standard functions RE, IM, CMPLX
  • standard functions INT, LFLOAT, MAX, MIN, LENGTH
  • standard functions can also be used in constant expressions within the declaration section, including real arithmetic, complex functions, SYSTEM.MAKEADR, and SYSTEM.CAST
  • record- and array value constructors with BY-count in declaration section and statement part
  • multi-dimensional open arrays
  • empty records
  • module SYSTEM: new address arithmetic functions ADDADR, SUBADR, DIFADR, as well as MAKEADR, ROTATE, SHIFT, type LOC (=BYTE) and constants LOCSPERBYTE, LOCSPERWORD, BITSPERLOC
  • generic real literal type RR: constant real literals and expressions are evaluated with longreal precision and are expression compatible with REAL and LONGREAL
  • generic complex literal type CC: constant complex numbers and expressions are evaluated with longcomplex precision and are expression compatible with COMPLEX and LONGCOMPLEX
  • any predeclared pervasive indentifier, such as for example REAL, LONGREAL, FLOAT, NIL can be redeclared
• additional ISO Modula-2 library modules to support mathematical functions, string conversions, and I/O for all VAX and Alpha floating point types
• additional ISO Modula-2 library modules to support mathematical functions for complex data types
• allows Modula-2 and Oberon-2 modules in the same project:
  • ISO Modula-2 library can be used in Oberon-2
  • common run-time system for Modula-2 and Oberon-2
  • when being called from Oberon-2, Modula-2 ignores Oberon-2's type descriptor for variable procedure parameters of record type and passes NIL vice versa
• Oberon-2 (A2O) also generates 64 bit load files for the AlphaOberon System (AOS)
  

Press releases, data sheets, and implementation notes

OpenVMS Alpha Modula-2 press release: MaX
OpenVMS Alpha Oberon-2 press release: A2O
OpenVMS Alpha & VAX Modula-2 data sheet (22KB html)
OpenVMS Alpha & VAX Oberon-2 data sheet (23KB html)
OpenVMS Alpha Modula-2 & Oberon-2 implementation notes (63KB html) (latest edition 20-Mar-1999)
OpenVMS Alpha Modula-2 & Oberon-2 The Year 10000 problem
OpenVMS Alpha Oberon-2 Compiler 64 bit address extension (A2O V3.0)
OpenVMS Alpha Modula-2 Compiler 64 bit address extension (MaX V5.0)
OpenVMS Alpha Modula-2 Application program migration from 32 bit to 64 bit (MaX V5)
OpenVMS Alpha Modula-2 64 bit Virtual Regions with Modula-2 on OpenVMS Alpha
Test Programs for the ISO Modula-2 Standard I/O Library

Oberon-2 LALR-Syntax
What is a 64 bit Compiler? (latest edition 20-Mar-1999)
Was ist Oberon-2? (a short Oberon-2 language description in German)

All data is subject to change without notice.

[ Home | Legal | Alpha_Oberon_System | ModulaTor | Bibliography | Oberon[-2]_links | Modula-2_links | General interesting book recommendations ]