VLISP A Verified Implementation of Scheme

The VLISP project showed how to produce a comprehensively verified implemen- tation for a programming language, namely Scheme [4, 15). Some of the major elements in this verification were: * The proof was based on the Clinger-Rees denotational semantics of Scheme given in [15). Our goal was to produce a "e;warts-and-all"e; verification of a real language. With very few exceptions, we constrained ourselves to use the se- mantic specification as published. The verification was intended to be rigorous, but. not. complet.ely formal, much in the style of ordinary mathematical discourse. Our goal was to verify the algorithms and data types used in the implementat.ion, not their embodiment. in code. See Section 2 for a more complete discussion ofthese issues. Our decision to be faithful to the published semantic specification led to the most difficult portions ofthe proofs; these are discussed in [13, Section 2.3-2.4). * Our implementation was based on the Scheme48 implementation of Kelsey and Rees [17). This implementation t.ranslates Scheme into an intermediate-level "e;byte code"e; language, which is interpreted by a virtual machine. The virtual machine is written in a subset of Scheme called PreScheme. The implementationissufficient.ly complete and efficient to allow it to bootstrap itself. We believe that this is the first. verified language implementation with these properties.

The VLISP project showed how to produce a comprehensively verified implemen- tation for a programming language, namely Scheme [4, 15). Some of the major elements in this verification were: * The proof was based on the Clinger-Rees denotational semantics of Scheme given in [15). Our goal was to produce a "e;warts-and-all"e; verification of a real language. With very few exceptions, we constrained ourselves to use the se- mantic specification as published. The verification was intended to be rigorous, but. not. complet.ely formal, much in the style of ordinary mathematical discourse. Our goal was to verify the algorithms and data types used in the implementat.ion, not their embodiment. in code. See Section 2 for a more complete discussion ofthese issues. Our decision to be faithful to the published semantic specification led to the most difficult portions ofthe proofs; these are discussed in [13, Section 2.3-2.4). * Our implementation was based on the Scheme48 implementation of Kelsey and Rees [17). This implementation t.ranslates Scheme into an intermediate-level "e;byte code"e; language, which is interpreted by a virtual machine. The virtual machine is written in a subset of Scheme called PreScheme. The implementationissufficient.ly complete and efficient to allow it to bootstrap itself. We believe that this is the first. verified language implementation with these properties.