November 2019 - java-devroom - lists.fosdem.org

Talk Proposal: Taming Metaspace
by Thomas Stüfe 23 Nov '19

23 Nov '19

* Title: Taming Metaspace: a look at the machinery, and a proposal for a better one * Abstract: When examining memory footprint of a JVM process, the delta between Java heap usage and actual working set size can be surprisingly large. The JVM uses off-heap memory for a lot of things: thread stacks, compiler arenas, code heap, byte buffers, GC control... however, one of the largest consumers of off-heap memory can be class metadata. Class metadata are stored in Metaspace, which includes the Compressed Class Space. The talk will explore what Metaspace actually is and what is stored there; describe the architecture of the Metaspace allocator and the Compressed Class Space; how it interacts with the GC; how it is sized. We will highlight waste areas and demonstrate how to use jcmd to examine Metaspace internals. The current implementation of the Metaspace allocator suffers from a number of shortcomings. They can manifest in excessive waste and a certain "clinginess" - an unwillingness to let go of unused memory. At SAP, we developed an improved version which is more frugal with memory and provides a much improved elasticity. So the second part of this talk will concentrate on our new implementation. We will highlight the differences to the old architecture, demonstrate advantages and examine how it works. * Recording me on audio and/or video acceptable under a CC-BY-2.0 license * Brief Bio Thomas is a JVM developer at SAP and a long time OpenJDK committer. He has worked with JVMs for the last fifteen years. * Microblog URL twitter.com/tstuefe * Blog URL stuefe.de

1 0

Talk proposal: Java bytecode verification. When, how, or turn it off maybe?
by Nikita Lipsky 22 Nov '19

22 Nov '19

Hi all, The talk may be interesting considering deprecation of -Xverify:none in JDK 13 (https://bugs.java.com/bugdatabase/view_bug.do?bug_id=JDK-8214719) - people should know why "-Xverify:none" is so dangerous. Title: Java bytecode verification. When, how, or turn it off maybe? Abstract: Today's Java developers are increasingly relying on libraries that generate Java bytecode at run time to effectively implement various tricks that are hard or impossible to do in pure Java. However, while the javac compiler is guaranteed to produce correct Java bytecode, shifting down to the bytecode level requires that you keep track of its correctness yourself. Otherwise, there is a chance to get a j.l.VerifyError at run time upon the loading of a generated class, because the JVM passes all classes through the Java bytecode verifier to enforce strict correctness requirements. Therefore, knowing just the semantics of Java bytecode instructions is often not enough to properly synthesize classes - you also need to know how the Java bytecode verifier works, what bytecode it deems correct, and what it does not. In this talk, we will learn what mission the Java bytecode verifier carries out in the JVM, when and how it works, whether it may affect the performance of your application, and why turning it off is so dangerous. Recording: ok Bio: Nikita works for Huawei on JVMs, compilers and new programming languages. Before Huawei, Nikita had been the key figure in the Excelsior JET project for over twenty years. Excelsior JET was a complete Java SE implementation (JVM) with an AOT compiler, and Nikita took part in almost all activities pertaining to that project, from core JVM development to product management and support. In particular, he is the author of OSGi support at the JVM level, the Java Runtime Slim-down technology - Java SE modularization implemented in Excelsior JET since 2007, both Java bytecode verifiers, Spring Boot support, and many other features. Nikita

1 0

Talk proposal: Shenandoah 2.0
by Roman Kennke 22 Nov '19

22 Nov '19

I'd like to propose the following talk for the Java DevRoom: Title: Shenandoah 2.0 Abstract: Shenandoah GC landed in JDK12 about a year ago, giving OpenJDK another low-pause garbage collector. It has undergone substantial changes since then. Specifically we have a new barrier scheme, and have eliminated the extra forwarding pointer word per object, thus substantially reducing memory footprint. After giving a general introduction to OpenJDK GC landscape and Shenandoah GC, this talk focuses on those recent changes in Shenandoah and what's in it for you. Bio: Roman Kennke is Prinipal Software Engineer at Red Hat, where he worked and works on OpenJDK and several related projects, including Thermostat, the Zero and Shark ports of OpenJDK, and currently on the Shenandoah GC. In past lifes, he worked at aicas (realtime Java VM), Sun Microsystems (Webstart, Swing) and JP Morgan.

1 0

The sinous path to Valhalla
by Remi Forax 22 Nov '19

22 Nov '19

Hi all, want to know more about the OpenJDK Valhalla project, about value^H^H^H^H^H inline types or to why we are 5 years in the making and still no JEP, this talk is for you :) title: The sinous path to Valhalla abstract: When Java was created at the beginning of the 90s, the cost of a pointer dereference and the cost of addition were roughly similar. Nowadays, we have hardware with hundreds of gigs of RAM, the cost of a pointer dereference is at least 2 order magnitude the cost of an addition, so the Java model of every object is a reference to an address in RAM is not fully aligned anymore with the current state of the hardware. It's time for a little lifting. During this session, I will tell the history of the Valhalla project, the choices that have been made, the breakthrough we had, what still needs to be done and when the project will land in the main OpenJDK branch. This talk is for the courageous that are not afraid about bytecode representation and work in progress details. Audio & Video recording: ok Short Bio: Remi Forax, Researcher, Teacher at Paris East University. One of the fathers of the invokedynamic bytecode, Java Champion that brought you lambdas, modules (yes i know), records, and more. Github repo: github.com/forax cheers, Rémi

1 0

Talk proposal - OpenArk Compiler: Generating binaries from Java source
by Bernhard Rosenkraenzer 21 Nov '19

21 Nov '19

Title: OpenArk Compiler: Generating binaries from Java source Abstract (at least two paragraphs): Huawei has recently announced a new compiler that compiles Java source code to native machine code: https://www.xda-developers.com/huawei-ark-compiler-android-app-performance/ We've taken a look at what's there already and where it's headed. This talk will be about what's there already, what's still needed, and various ways in which this can be useful for Free Java projects. If possible describe why Free Java is helpful or perhaps even essential to your project: The project is a Free Java compiler - it provides a new way to run Free Java applications and couldn't exist without all the other Free Java components. What the audience will learn from the talk: What OpenArk can do, how to use it, and where it's headed. How can people help out with the project, what feedback are you looking for? There's a number of things that can be done to improve the compiler, including but not limited to helping add support for different target platforms (current main target is Android on aarch64). Also, more testing is needed. Could use feedback on what other features the Free Java community would like to see in the compiler. Recording me on audio and/or video: acceptable under a CC-BY-2.0 license (DEFAULT) Brief Bio: Bernhard "Bero" Rosenkränzer is one of the main developers of OpenMandriva, and the current president of the OpenMandriva Association. He currently runs a development and consultancy company called LinDev. Previous work places include Linaro, Red Hat, MandrakeSoft and lesser known startups. He has committed code all over the Linux stack, ranging from the kernel to the KDE Plasma desktop and Android. He has spoken at many Linux/Open Source conferences, including Linux Plumbers Conference, Embedded Linux Conference, Open Source Summit, Android Builders Summit, LinuxTag and Linux Piter. Microblog URL: Blog URL: I don't currently have my own (Micro)blogs, but occasionally write guest posts on other people's.

1 0

Talk Proposal: How I learned to love my Dragon: The adventures of transpiling Java bytecode to native code for the CrossMobile platform
by Panayotis Katsaloulis 20 Nov '19

20 Nov '19

* Title How I learned to love my Dragon: The adventures of transpiling Java bytecode to native code for the CrossMobile platform * Abstract Using Java as a high level language and converting to low native code is not an easy task. You have to fight with mismatching issues like GC and memory management, different APIs, hardware access, and a common simple high-level interface. On top of this, CrossMobile is a tool that addresses iOS, Android, and Desktop Java at once, which leads to issues such as different UI experience, synchronous vs asynchronous methods, C-styled varargs, i18n, complex constructs like web views and maps, and the list goes on. This talk will cover how these problems were solved and the experience gained through the procedure, what mistakes to avoid and how a high-level language like Free Java was a good strategic decision to accomplish this task. As a closing remark will be where to go from now and the possibility to support any JVM based language such as Clojure or Kotlin, while keeping the developers happy and platform-agnostic. * Recording me on audio and/or video acceptable under a CC-BY-2.0 license * Brief Bio Dr. Panayotis Katsaloulis is one of the first and awarded Android and iOS developers, long-term Swing user, and an Open Source evangelist. He was the vice president of Greek LUG for many years and helped the Open Source and Java ecosystem in Greece to grow. Right now he is the CEO of CrossMobile tech and the core developer of the CrossMobile platform, which recently has been open sourced. * URLs https://github.com/crossmob/CrossMobile https://crossmobile.tech/ http://panayotis.com/

1 0

Talk proposal for Java Developer Room at FOSDEM 2020
by Nikola Grcevski 19 Nov '19

19 Nov '19

Hi there, I'd like to submit the following talk proposal for the Java Developer Room: Title: Reducing OpenJDK Java Garbage Collection times with stack allocation Abstract: In this talk we'll explore ways that the JVM can reduce the object allocation rate of Java programs automatically by performing stack allocation of objects that are known to be local to a method, or in compiler terms non-escaping. The discussion is focused on employing the escape analysis optimization in the OpenJDK Hotspot C2 compiler to determine which Java objects can be stack allocated, and how this optimization can reduce pressure on the Java JVM garbage collectors. We'll show some results on how various real world applications can benefit from such optimizations and describe the methodology of how we prototyped this in OpenJDK. Our work is only in prototype state at this moment and we are looking for more data to understand how broadly applicable this optimizations is. This work wouldn't be possible without free open source access to Java. Recording me on audio and/or video: 1. Acceptable under CC-BY-2.0 license Brief Bio: I started coding when I was 9 on the now legendary ZX Spectrum (hello Black Mirror - Bandersnatch), but got really seriously into it with C and assembly on early x86 Intel processors. I have been working on JVM (JIT compilation and GC) technologies on and off for almost 20 years. I did most of my VM work on OpenJ9 Testarossa JIT compiler and at the moment I'm VM engineer at Microsoft working on OpenJDK. I live in Toronto, Canada with my wife, 2 kids and a cat and like most Canadians I'm true indoor enthusiast. Microblog URL: https://twitter.com/vivaboredom Blog URL: I'm not much of a blog writer (yet!), so here's my github profile instead: https://github.com/grcevski

1 0

Talk Proposal: Discover Dependency License Information with ClearlyDefined
by Jeff Mendoza 19 Nov '19

19 Nov '19

Title: Discover Dependency License Information with ClearlyDefined Abstract: Complying with license obligations can incur a lot of hurdles. This results in developers skipping compliance or failing to achieve correct compliance. To compound this, package authors sometimes fail to describe the license of their package clearly or omit license information of included files. ClearlyDefined is a community curated repository of discovered license information for Maven packages, among other types. This talk will cover: the importance of the license obligations of the dependencies of your Java package, tool, or application. How to discover the license information of those packages. And demonstrate some Maven compatible tooling that allows automated license recordkeeping and notice file generation as a part of your CI system. Recording: acceptable under a CC-BY-2.0 license Brief Bio: Jeff Mendoza is a long time FOSS enthusiast. During the day, as a member of Microsoft’s Open Source Programs Office, he enables Microsoft engineers to engage with Open Source effectively. At night he plays games of various media and enjoys pushing the boundaries of self-hosting. About ClearlyDefined: https://docs.clearlydefined.io/ Thanks! Jeff

1 0

Talk Proposal
by Mike Milinkovich 18 Nov '19

18 Nov '19

Dear all, Thanks for your consideration. Title: Free at Last! The Tale of Jakarta EE Abstract (at least two paragraphs) In September 2017 Oracle announced that it would be migrating governance of the Java EE platform to the Eclipse Foundation, the home of MicroProfile. Two years later Jakarta EE 8 shipped, signaling the successful completion of that move. As a result, Free Java has a new home for a significant piece of the Java ecosystem. A home which is purely open source, vendor neutral, and community led. This talk will be about how the long and painful journey from Java EE to Jakarta EE unfolded. But more importantly it will focus on how the new Jakarta EE community works, and how there is a new, open, specification process for Java APIs (other than SE) that is available for the community. We are looking forward to welcoming many of those interested in Free Java to participate in driving new innovation in Java APIs for cloud and other exciting use cases. Recording me on audio and/or video acceptable under a CC-BY-2.0license (YES) Brief Bio: Mike Milinkovich is the Executive Director of the Eclipse Foundation. Microblog URL: @mmilinkov @ Twitter Blog UR: https://eclipse-foundation.blog/ Mike Milinkovich mike.milinkovich(a)eclipse-foundation.org (m) +1.613.220.3223

1 0

Project Loom - Talk proposal
by Andrew Haley 01 Nov '19

01 Nov '19

Title: Project Loom: Advanced concurrency for fun and profit Abstract: Project Loom, an OpenJDK project, is "intended to explore, incubate and deliver Java VM features and APIs built on top of them for the purpose of supporting easy-to-use, high-throughput lightweight concurrency and new programming models on the Java platform." These feature include Lightweight Threads, delimited continuations, and tail-call elimination. The speaker, a Project Loom team member, will describe the project in depth, in particular the gnarly details of how coroutine and continuation scheduling mechanism works, and a new feature, Scoped Locals. Recording me on audio and/or video: I'm cool with that. Bio: Andrew Haley is a Distinguished Engineer at Red Hat. He leads the Java team there, the OpenJDK 8 and 11 Updates projects and was a co-author of the 64-bit OpenJDK ARM port. -- Andrew Haley (he/him) Java Platform Lead Engineer Red Hat UK Ltd. <https://www.redhat.com> https://keybase.io/andrewhaley EAC8 43EB D3EF DB98 CC77 2FAD A5CD 6035 332F A671

1 0