A machine-independent microprogram development system
- Authors: Ward, Michael John
- Date: 1987 , 2013-03-11
- Subjects: Microprogramming
- Language: English
- Type: Thesis , Masters , MSc
- Identifier: vital:4581 , http://hdl.handle.net/10962/d1003738 , Microprogramming
- Description: The aims of this project are twofold. They are firstly, to implement a microprogram development system that allows the programmer to write microcode for any microprogrammable machine, and secondly, to build a microprogrammable machine, incorporating the user friendliness of a simulator, while still providing the 'hands on' experience obtained actual hardware. Microprogram development involves a two stage process. The first step is to describe the target machine, using format descriptions and mnemonic-based template definitions. The second stage involves using the defined mnemonics to write the microcodes for the target machine. This includes an assembly phase to translate the mnemonics into the binary microinstructions. Three main components constitute the microprogrammable machine. The Arithmetic and Logic Unit (ALU) is built using chips from Advanced Micro Devices' Am29ØØ bit-slice family, the action of the Microprogram Control Unit (MCU) is simulated by software running on an IBM Personal Computer, and a section of the IBM PC's main memory acts as the Control Store (CS) for the system. The ALU is built on a prototyping card that plugs into one of the slots on the IBM PC's mother board. A hardware simulator program, that produces the effect of the ALU, has also been developed. A small assembly language has been developed using the system, to test the various functions of the system. A mini-assembler has also been written to facilitate assembly of the above language. A group of honours students at Rhodes University tested the microprogram development system. Their ideas and suggestions have been tabulated in this report and some of them have been used to enhance the system's performance. The concept of allowing 'inline' microinstructions in the macroprogram is also investigated in this report and a method of implementing this is shown.
- Full Text:
- Date Issued: 1987
- Authors: Ward, Michael John
- Date: 1987 , 2013-03-11
- Subjects: Microprogramming
- Language: English
- Type: Thesis , Masters , MSc
- Identifier: vital:4581 , http://hdl.handle.net/10962/d1003738 , Microprogramming
- Description: The aims of this project are twofold. They are firstly, to implement a microprogram development system that allows the programmer to write microcode for any microprogrammable machine, and secondly, to build a microprogrammable machine, incorporating the user friendliness of a simulator, while still providing the 'hands on' experience obtained actual hardware. Microprogram development involves a two stage process. The first step is to describe the target machine, using format descriptions and mnemonic-based template definitions. The second stage involves using the defined mnemonics to write the microcodes for the target machine. This includes an assembly phase to translate the mnemonics into the binary microinstructions. Three main components constitute the microprogrammable machine. The Arithmetic and Logic Unit (ALU) is built using chips from Advanced Micro Devices' Am29ØØ bit-slice family, the action of the Microprogram Control Unit (MCU) is simulated by software running on an IBM Personal Computer, and a section of the IBM PC's main memory acts as the Control Store (CS) for the system. The ALU is built on a prototyping card that plugs into one of the slots on the IBM PC's mother board. A hardware simulator program, that produces the effect of the ALU, has also been developed. A small assembly language has been developed using the system, to test the various functions of the system. A mini-assembler has also been written to facilitate assembly of the above language. A group of honours students at Rhodes University tested the microprogram development system. Their ideas and suggestions have been tabulated in this report and some of them have been used to enhance the system's performance. The concept of allowing 'inline' microinstructions in the macroprogram is also investigated in this report and a method of implementing this is shown.
- Full Text:
- Date Issued: 1987
CSP-i : an implementation of CSP
- Authors: Wrench, Karen Lee
- Date: 1987 , 2013-03-08
- Subjects: Synchronization--Computers , Programming languages (Electronic computers)
- Language: English
- Type: Thesis , Masters , MSc
- Identifier: vital:4579 , http://hdl.handle.net/10962/d1003124 , Synchronization--Computers , Programming languages (Electronic computers)
- Description: CSP (Communicating Sequential Processes) is a notation proposed by Hoare, for expressing process communication and synchronization. Although this notation has been widely acclaimed, Hoare himself never implemented it as a computer language. He did however produce the necessary correctness proofs and subsequently the notation has been adopted (in various guises) by the designers of other concurrent languages such as Ada and occam. Only two attempts have been made at a direct and precise implementation of CSP. With closer scrutiny, even these implementations are found to deviate from the specifications expounded by Hoare, and in so doing restrict the original proposal. This thesis comprises two main sections. The first of these includes a brief look at the primitives of concurrent programming, followed by a comparative study of the existing adaptations of CSP and other message passing languages. The latter section is devoted to a description of the author's attempt at an original implementation of the notation. The result of this attempt is the creation of the CSP-i language and a suitable environment for executing CSP-i programs on an IBM PC. The CSP-i implementation is comparable with other concurrent systems presently available. In some aspects, the primitives featured in CSP-i provide the user with a more efficient and concise notation for expressing concurrent algorithms than several other message-based languages, notably occam. , KMBT_363 , Adobe Acrobat 9.53 Paper Capture Plug-in
- Full Text:
- Date Issued: 1987
- Authors: Wrench, Karen Lee
- Date: 1987 , 2013-03-08
- Subjects: Synchronization--Computers , Programming languages (Electronic computers)
- Language: English
- Type: Thesis , Masters , MSc
- Identifier: vital:4579 , http://hdl.handle.net/10962/d1003124 , Synchronization--Computers , Programming languages (Electronic computers)
- Description: CSP (Communicating Sequential Processes) is a notation proposed by Hoare, for expressing process communication and synchronization. Although this notation has been widely acclaimed, Hoare himself never implemented it as a computer language. He did however produce the necessary correctness proofs and subsequently the notation has been adopted (in various guises) by the designers of other concurrent languages such as Ada and occam. Only two attempts have been made at a direct and precise implementation of CSP. With closer scrutiny, even these implementations are found to deviate from the specifications expounded by Hoare, and in so doing restrict the original proposal. This thesis comprises two main sections. The first of these includes a brief look at the primitives of concurrent programming, followed by a comparative study of the existing adaptations of CSP and other message passing languages. The latter section is devoted to a description of the author's attempt at an original implementation of the notation. The result of this attempt is the creation of the CSP-i language and a suitable environment for executing CSP-i programs on an IBM PC. The CSP-i implementation is comparable with other concurrent systems presently available. In some aspects, the primitives featured in CSP-i provide the user with a more efficient and concise notation for expressing concurrent algorithms than several other message-based languages, notably occam. , KMBT_363 , Adobe Acrobat 9.53 Paper Capture Plug-in
- Full Text:
- Date Issued: 1987
The P.R.O. expert system shell
- Authors: Bradshaw, John
- Date: 1987 , 2013-04-03
- Subjects: Expert systems (Computer science)
- Language: English
- Type: Thesis , Masters , MSc
- Identifier: vital:4617 , http://hdl.handle.net/10962/d1006302 , Expert systems (Computer science)
- Description: This thesis reports the research which led to the development of the P.R .O. Expert System Shell. The P.R.O . System is primarily, though not exclusively , designed for use in ecological domains. In the light of two specific expert systems, The RCS (River Conservation Status) and the Aquaculture Systems, which were developed as part of this research, a number of areas of importance have been identified. The most significant of these is the need to handle uncertainty effectively. The style of knowledge representation to be implemented also plays an important role. After consulting the relevant literature and the available microcomputer expert system shells, a number of ideas have been included in the P.R.O. System. The P.R.O . System is a backward chaining, production system based expert system shell. It embodies a simple but effective method of handling uncertainty. An important feature of this method is that it takes cognizance of the different relative importances of the conditions which need to be satisfied before a conclusion can be reached. The knowledge base consists of more than rules and questions. It also contains meta-knowledge, which is used by the inference engine. The P.R.O. System has been designed to be of practical use. Its strongest recommendations are therefore, that the two non-trivial systems which have been implemented in it, have been accepted by the experts and their peers as systems which produce good, accurate answers . , KMBT_363 , Adobe Acrobat 9.53 Paper Capture Plug-in
- Full Text:
- Date Issued: 1987
- Authors: Bradshaw, John
- Date: 1987 , 2013-04-03
- Subjects: Expert systems (Computer science)
- Language: English
- Type: Thesis , Masters , MSc
- Identifier: vital:4617 , http://hdl.handle.net/10962/d1006302 , Expert systems (Computer science)
- Description: This thesis reports the research which led to the development of the P.R .O. Expert System Shell. The P.R.O . System is primarily, though not exclusively , designed for use in ecological domains. In the light of two specific expert systems, The RCS (River Conservation Status) and the Aquaculture Systems, which were developed as part of this research, a number of areas of importance have been identified. The most significant of these is the need to handle uncertainty effectively. The style of knowledge representation to be implemented also plays an important role. After consulting the relevant literature and the available microcomputer expert system shells, a number of ideas have been included in the P.R.O. System. The P.R.O . System is a backward chaining, production system based expert system shell. It embodies a simple but effective method of handling uncertainty. An important feature of this method is that it takes cognizance of the different relative importances of the conditions which need to be satisfied before a conclusion can be reached. The knowledge base consists of more than rules and questions. It also contains meta-knowledge, which is used by the inference engine. The P.R.O. System has been designed to be of practical use. Its strongest recommendations are therefore, that the two non-trivial systems which have been implemented in it, have been accepted by the experts and their peers as systems which produce good, accurate answers . , KMBT_363 , Adobe Acrobat 9.53 Paper Capture Plug-in
- Full Text:
- Date Issued: 1987
- «
- ‹
- 1
- ›
- »