This paper describes our design for an optimizer that we incorporated into the Esterel compiler.

/or password checks. In addition, an application code may need to be restructured to work under such a memory system. In this paper, we propose a compiler-directed strategy to generate code for a secure memory based embedded architecture. The idea is to let the programmer mark certain data elements, called

manually. We present results that suggest this is a viable approach to high-performance XML parsing.

type encodes programmerdefined types that Mockingbird is expected to leave alone. A Capability type is used to model 7 Character repertoire = "rep-name" Record member TYPE member TYPE Choice member TYPE member TYPE Sequence countrange = range-spec TYPE Capability special-repertoire = name Integer range = range-spec Real exponent = range-spec Any Miscellaneous Primitive Types Structuring Types Communication Channels Type Reference TypeRef Abstract name = type-name-string precision = precision-spec ref to MockSL type ref to MockSL type Figure 6: MockSL type constructors remote references. A TypeRef type is used to express recursion or type inclusion when one type is completely included in the definition of another. A Record type constructs heterogeneous aggregates and a Sequence type constructs homogeneous aggregates. Finally, a Choice type constructs alternative types (as in the variant or union types of Pascal and C). Because the MockSL type system is ...

. Existing techniques fail in such cases, bringing down the performance. In this paper, we extend the framework that exists today, to the case of irregular access. We provide a clear and precise compiler-based technique for analyzing irregular array-access, and efficiently mapping such arrays to the scratch

Abstract. In this paper, we evaluate two different approaches for the compiler-based parallelisation ofa C program for MPEG-2 decoding. The first approach experiments with a commercial uto-parallelising compiler for exploiting coarse-grain parallelism on a Silicon Graphics Power Chal-lenge, while

Intra-iteration data reuse occurs when multiple array references exhibit data reuse in a single loop iteration. An optimizing compiler can exploit this reuse by clustering (in the loop body) array references with data reuse as much as possible. This reduces the number of intervening references

of providing a critical level of timely service even in the presence of software or hardware faults. We argue that providing hardware fault-tolerance for real-time systems requires support from the language, compiler, and environment. We suggest an integrated approach to providing this support. The key aspects

manually. We present results that suggest this is a viable approach to high-performance XML parsing. 1

systems. Currently, validation is considered expensive, but we posit that schema information can actually be used to speedup parsing. This paper develops a framework for such parsing, called schema-specific parsing, and presents preliminary results that show this is indeed a viable approach to high