Path: g2news1.google.com!news1.google.com!news.glorb.com!nntp-server.pubsub.com!netnews.sdu.edu.cn!USTC From: Galoisx....@bbs.ustc.edu.cn (昆虫博士) Newsgroups: cn.bbs.comp.unix Subject: DragonFly BSD 1.0 released! Date: 13 Jul 2004 05:10:55 GMT Organization: 瀚海星云 Message-ID: <4DDE3T$Ci3@bbs.ustc.edu.cn> X-Filename: Unix/M.1089695405.A Lines: 93 DragonFly-1.0 RELEASED! 12 July 2004 http://www.dragonflybsd.org/main/ One year after starting the project as a fork off the FreeBSD-4.x tree, the DragonFly Team is pleased to announce our 1.0 release! We've made remarkable progress in our first year. We have replaced nearly all of the core threading, process, interrupt, and network infrastructure with DragonFly native subsystems. We have our own MP-friendly slab allocator, a Light Weight Kernel Threading (LWKT) system that is separate from the dynamic userland scheduler, a fine-grained system timer abstraction for kernel use, a fully integrated light weight messaging system, and a core IPI (Inter Processor Interrupts) messaging system for inter-processor communications. We have managed to retain 4.x's vaunted stability throughout the development process, despite ripping out and replacing major subsystems, and we have a demonstratively superior coding model which is both UP (Uni-Processor) and MP (Multi-Processor) friendly and which is nearly as efficient on UP systems as the original 4.x UP-centric code is on UP systems. We have made excellent progress bringing in those pieces from FreeBSD, NetBSD, and OpenBSD that fit our model. For example, NEWBUS/BUS_DMA, the USB infrastructure, RCNG (next generation system startup infrastructure), and so forth. We have made an excellent start on reformulating the build and release infrastructure including an excellent new system installer which, while still in its infancy for the 1.0 release, has been coded in a manner that will allow us to greatly improve and expand its capabilities in coming months. We have done so much that it cannot all be listed here. Please check out the Diary for technical details. The two largest user-visible subsystems that still have major work pending are the userland threading and ports/packages subsystems. People will find that the DragonFly-1.0 release is still using the old 4.x pthreads model, and at the moment we are relying on the FreeBSD ports tree with DragonFly specific overrides for third party application support... about as severe a hack as it is possible to have. These two stop-gap items will be at the forefront of the work for the next year, along with a major move to start removing the BGL (Big Giant Lock, also known as the MP lock) from code inherited from 4.x, threading the VFS (Virtual File System) subsystem (the network subsystem is already threaded as of 1.0), and implementing asynchronously messaged system calls. And that is just the tip of the iceberg, for we will be achieving far more in the coming year! What is DragonFly BSD? DragonFly is an operating system and environment designed to be the logical continuation of the FreeBSD-4.x OS series. These operating systems belong in the same class as Linux in that they are based on UNIX ideals and APIs. DragonFly is a fork in the path, so to speak, giving the BSD base an opportunity to grow in an entirely new direction from the one taken in the FreeBSD-5 series. It is our belief that the correct choice of features and algorithms can yield the potential for excellent scalability, robustness, and debuggability in a number of broad system categories. Not just for SMP or NUMA, but for everything from a single-node UP system to a massively clustered system. It is our belief that a fairly simple but wide-ranging set of goals will lay the groundwork for future growth. The existing BSD cores, including FreeBSD-5, are still primarily based on models which could at best be called 'strained' as they are applied to modern systems. The true innovation has given way to basically just laying on hacks to add features, such as encrypted disks and security layering that in a better environment could be developed at far less cost and with far greater flexibility. We also believe that it is important to provide API solutions which allow reasonable backwards and forwards version compatibility, at least between userland and the kernel, in a mix-and-match environment. If one considers the situation from the ultimate in clustering... secure anonymous system clustering over the internet, the necessity of having properly specified APIs becomes apparent. Finally, we believe that a fully integrated and feature-full upgrade mechanism should exist to allow end users and system operators of all walks of life to easily maintain their systems. Debian Linux has shown us the way, but it is possible to do better. DragonFly is going to be a multi-year project at the very least. Achieving our goal set will require a great deal of groundwork just to reposition existing mechanisms to fit the new models. The goals link will take you to a more detailed description of what we hope to accomplish. -- ============================================================= || 心中的神灵Evariste Galois熟视无睹 || ------------------------------------------------------------- || In daytime, I am a Penguin. || z...@mail.ustc.edu.cn || || At night, I am a Daemon. || G2 Project. || ============================================================= ※ 来源:·瀚海星云 bbs.ustc.edu.cn·[FROM: 210.45.124.197]