Eclipse OMR / OMR Technical Talks

Vijay Sundaresan describes the OMR data flow analyses in the OMR compiler.

Vijay Sundaresan gives an overview of the value propagation optimization in the OMR compiler. There are no slides, just useful information!

Currently, Eclipse OpenJ9 uses OMR’s Test & Test & Set (TATAS) locks a.k.a. spinlocks with compare and swap (CAS) for synchronization, which are known to be unfair. TATAS locks collapse on massively parallel systems during high lock contention where many threads attempt to acquire a lock simultaneously. This talk will cover the following:
1) Does transitioning to scalable locks, such as the Mellor-Crummey & Scott (MCS) queue-based spinlock, in OpenJ9 resolve the TATAS bottleneck?2) Do features, such as lock cohorting, concurrency restriction, transactional lock elision (TLE) and scalable statistics counters, help further improve locking performance in OpenJ9?

Presentation on how to leverage Eclipse OMR JitBuilder API to convert LLVM IR to OMR IL to generate optimal native machine instructions across a variety of platforms OMR supports.

Eclipse OMR/OpenJ9's software-based Concurrent Scavenge GC policy.

Improved JVM debugging mode based on OSR

Profile driven optimization is very useful in improving runtime performance especially in Just-in-Time compilations where application are compiled while running. In this talk, Rahil will first give an overview on the different profiling mechanisms including those used in OpenJ9 and discuss their advantages and disadvantages. He will then explain the low-overhead patchable runtime profiling mechanism called JProfiling, how it works and its advantages over the other mechanisms.

Ever wonder why IBM Z mainframes have been such an attractive and successful platform for modern enterprises? In this talk, Iris will discuss IBM Z’s strengths in performance, scalability, security, isolation and RAS.

Thanks to @nbhuiyan who fixed the first 12 minutes of audio that were missing from the initial video! Here is the newly re-recorded and stitched together Compiler Vitality Talk -- Code Generators and Much More (Part II).

In this talk, Julian will give an overview of the JIT code generator framework for the platforms we support, namely X, P, and Z. In particular, he will discuss what concepts/code is currently being shared among the code generators, and what is not. He will compare the architectures, reason about the philosophy of the respective construction, and how the modern micro-processor differences result in differences in our JIT code generators.

This talk gives a high-level overview of the extensible class design in the Eclipse OMR compiler technology. It describes the motivation behind it, how to architect classes and enums for extensibility, and provides abstract and concrete examples.

This is a quick tour of the compiler technology in Eclipse OMR in action. It provides an overview of the overall architecture, IL generation, high-level optimizer, and code generator all in the context of OpenJ9.