Compiling for EDGE Architectures

Publisher Website
Google Scholar
Abstract
Explicit Data Graph Execution (EDGE) architectures of- fer the possibility of high instruction-level parallelismwith energy efficiency. In EDGE architectures, the compiler breaks a program into a sequence of structured blocks that the hardware executes atomically. The instructions within each block communicate directly, instead of communicating through shared registers. The TRIPS EDGE architecture imposes restrictions on its blocks to simplify the microar- chitecture: each TRIPS block has at most 128 instructions, issues at most 32 loads and/or stores, and executes at most 32 register bank reads and 32 writes. To detect block com- pletion, each TRIPS block must produce a constant number of outputs (stores and register writes) and a branch deci- sion. The goal of the TRIPS compiler is to produce TRIPS blocks full of useful instructions while enforcing these co n- straints. This paper describes a set of compiler algorithms that meet these sometimes conflicting goals, including an algorithm that assigns load and store identifiers to maxi- mize the number of loads and stores within a block. We demonstrate the correctness of these algorithms in simu- lation on SPEC2000, EEMBC, and microbenchmarks ex- tracted from SPEC2000 and others. We measure speedup in cycles over an Alpha 21264 on microbenchmarks.

This publication has 21 references indexed in Scilit: