2024/10/22 by Padmaraman, Vedanth, Chilamkurthy, Sasank
#FOS: Computer and information sciences #Programming Languages (cs.PL)
paper · doi:10.48550/arxiv.2410.16690
Llama.lisp is a compiler framework intended to target offload processor backends such as GPUs, using intermediate representation languages (IRs) that are device-agnostic. The Llama.lisp IRs are formulated as S-expressions. This makes them easy to generate using higher level programming languages, which is one of the primary goals for Llama.lisp. The highest IR layer currently implemented in Llama.lisp is C-Lisp. In this paper, we describe the macro system developed for the Llama.lisp compiler framework. We show how we implemented FFI bindings as an example of this system.