Erven Rohou

Senior Researcher at Inria[Erven ROHOU]
Head of the PACAP Research Group

Research Interests

My current research interests include aspects of static and just-in-time compilation, as well as dynamic binary rewriting. I am particularly interested in the interaction of compiler optimizations with micro-architectural features and performance assessment.

More recently, we started investigating compilation techniques dedicated to non-volatile memory and intermittently powered systems.

We also look into how compilation techniques help security mechanisms.

Selected Publications

  • Erven Rohou, Bharath Narasimha Swamy, and André Seznec. Branch prediction and the performance of interpreters – don’t trust folklore. In: IEEE/ACM International Symposium on Code Generation and Optimization (CGO), 2015.
  • Simon Rokicki, Erven Rohou, Steven Derrien, Hybrid-DBT: Hardware/Software Dynamic Binary Translation Targeting VLIW, In: IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems (TCAD), 2018.
  • Erven Rohou, Sergei Dyshel, Dorit Nuzman, Ira Rosen, Kevin Williams, Albert Cohen, and Ayal Zaks, Speculatively Vectorized Bytecode, In Proceedings of 6th ACM International Conference on High-Performance and Embedded Architectures and Compilers (HiPEAC), 2011.
  • Rabab Bouziane, Erven Rohou, Abdoulaye Gamatié, Energy-efficient memory mappings based on partial WCET analysis and multi-retention time STT-RAM, In: Proceedings of the 26th International Conference on Real-Time Networks and Systems (RTNS), 2018.

Google Scholar seems to otherwise maintain a good publication list.

Recent Software

  • Tiptop a simple tool, very similar to the UNIX top utility, that displays the values of hardware performance counters associated to processes and threads. The goal is to make the collection of performance and bottleneck data as simple as possible, including simple installation and usage. No privilege is required, any user can run tiptop. Tiptop has been integrated in major Linux distributions, such as Fedora, Debian, Ubuntu. To give it a try:
    # dnf install tiptop
  • If-memo is a linker-based technique for enabling software memorizing of any dynamically linked pure function by function interception. Typically, this framework is useful to intercept the computationally expensive pure functions – the transcendental functions from the math library. Our technique does not need the availability of source code and thus can even be applied to commercial applications as well as applications with legacy codes. As far as users are concerned, enabling memoization is as simple as setting an environment variable. Our framework does not make any specific assumptions about the underlying architecture or compiler too-chains, and can work with a variety of current architectures.

Comments are closed.