Research
Embecosm relies on a heavy commitment to compiler research in order to stay at the forefront of modern compiler technology. Much of this is supported by Innovate UK, the British government’s innovation agency. Here you can find information about our main research projects, both complete and in-progress.
Pages in this section
SECURE
The Security Enhancing Compilation for Use in Real Environments (SECURE) project is an InnovateUK supported research program by Embecosm which started in July 2017. Its goal is to take the latest academic ideas for improving security of code, and provide practical reference implications in the main open source compilers, GCC and LLVM. The project has […]
GSO 2.0
The original GNU Superoptimizer (GSO) was pioneering, but has some limitations: it only handles arithmetic instructions, delivers a single result, generates impossible sequences, only supports a carry flag and has a very simple cost model. Its big advantage is that it is fast. Since GSO was created in 1991, there have been many advances in […]
AAP
An Altruistic Processor (AAP) was created to advance compiler technology for deeply embedded processors with a restricted register set and complex memory structures. The first version is documented in Embecosm Application Note 13. It is a 16-bit Harvard architecture with multiple 16-bit word addressed code memories, multiple byte addressed data memories and between 4 and […]
TSERO
2Total Software Energy Reduction and Optimization (TSERO) was an InnovateUK supported follow-on project between Embecosm, Allinea (now part of ARM), Concertim and STFC Daresbury Hartree Center, running from June 2015 to September 2017. The project aimed to apply the techniques developed in the MAGEEC and Superoptimization projects to compiling energy efficient code for high performance computing […]
Superoptimization
Compilers translate software into code executed by actual processors, but that translation is not always as efficient as desired, even when using traditional optimization techniques are used. Superoptimization attempts to find the the theoretically best translation of a block of code (in terms of code size, execution speed and energy efficiency). The first attempts were […]
MAGEEC
The Machine Guided Energy Efficient Compiler (MAGEEEC) project was an InnovateUK supported research program led by Embecosm in partnership with the University of Bristol from May 2013 to November 2014. Its goal was to make machine learning feasible in commercial compilers, specifically for generating energy efficient code on deeply embedded systems. The target was to […]