DragonFly BSD

DragonFly BSD is a free and open-source Unix-like operating system forked from FreeBSD 4.8. Matthew Dillon, an Amiga developer in the late 1980s and early 1990s and FreeBSD developer between 1994 and 2003, began working on DragonFly BSD in June 2003 and announced it on the FreeBSD mailing lists on 16 July 2003.

Dillon started DragonFly in the belief that the techniques adopted for threading and symmetric multiprocessing in FreeBSD 5 would lead to poor performance and maintenance problems. He sought to correct these anticipated problems within the FreeBSD project. Due to conflicts with other FreeBSD developers over the implementation of his ideas, his ability to directly change the codebase was eventually revoked. Despite this, the DragonFly BSD and FreeBSD projects still work together, sharing bug fixes, driver updates, and other improvements.

Intended as the logical continuation of the FreeBSD 4.x series, DragonFly has diverged significantly from FreeBSD, implementing lightweight kernel threads (LWKT), an in-kernel message passing system, and the HAMMER file system. Many design concepts were influenced by AmigaOS.

System design

Kernel

The kernel messaging subsystem being developed is similar to those found in microkernels such as Mach, though it is less complex by design. DragonFly's messaging subsystem has the ability to act in either a synchronous or asynchronous fashion, and attempts to use this capability to achieve the best performance possible in any given situation.

According to developer Matthew Dillon, progress is being made to provide both device input/output (I/O) and virtual file system (VFS) messaging capabilities that will enable the remainder of the project goals to be met. The new infrastructure will allow many parts of the kernel to be migrated out into userspace; here they will be more easily debugged as they will be smaller, isolated programs, instead of being small parts entwined in a larger chunk of code. Additionally, the migration of select kernel code into userspace has the benefit of making the system more robust; if a userspace driver crashes, it will not crash the kernel.

System calls are being split into userland and kernel versions and being encapsulated into messages. This will help reduce the size and complexity of the kernel by moving variants of standard system calls into a userland compatibility layer, and help maintain forwards and backwards compatibility between DragonFly versions. Linux and other Unix-like OS compatibility code is being migrated out similarly.

Threading

As support for multiple instruction set architectures complicates symmetric multiprocessing (SMP) support, DragonFly BSD now limits its support to the x86-64 platform. DragonFly originally ran on the x86 architecture, however as of version 4.0 it is no longer supported. Since version 1.10, DragonFly supports 1:1 userland threading (one kernel thread per userland thread), which is regarded as a relatively simple solution that is also easy to maintain. Inherited from FreeBSD, DragonFly also supports multi-threading.

In DragonFly, each CPU has its own thread scheduler. Upon creation, threads are assigned to processors and are never preemptively switched from one processor to another; they are only migrated by the passing of an inter-processor interrupt (IPI) message between the CPUs involved. Inter-processor thread scheduling is also accomplished by sending asynchronous IPI messages. One advantage to this clean compartmentalization of the threading subsystem is that the processors' on-board caches in symmetric multiprocessor systems do not contain duplicated data, allowing for higher performance by giving each processor in the system the ability to use its own cache to store different things to work on.

The ''LWKT'' subsystem is being employed to partition work among multiple kernel threads (for example in the networking code there is one thread per protocol per processor), reducing competition by removing the need to share certain resources among various kernel tasks.

Shared resources protection

In order to run safely on multiprocessor machines, access to shared resources (like files, data structures) must be serialized so that threads or processes do not attempt to modify the same resource at the same time. In order to prevent multiple threads from accessing or modifying a shared resource simultaneously, DragonFly employs critical sections, and serializing tokens to prevent concurrent access. While both Linux and FreeBSD 5 employ fine-grained mutex models to achieve higher performance on multiprocessor systems, DragonFly does not. Until recently, DragonFly also employed spls, but these were replaced with critical sections.

Much of the system's core, including the ''LWKT'' subsystem, the IPI messaging subsystem and the new kernel memory allocator, are lockless, meaning that they work without using mutexes, with each process operating on a single CPU. Critical sections are used to protect against local interrupts, individually for each CPU, guaranteeing that a thread currently being executed will not be preempted.

Serializing tokens are used to prevent concurrent accesses from other CPUs and may be held simultaneously by multiple threads, ensuring that only one of those threads is running at any given time. Blocked or sleeping threads therefore do not prevent other threads from accessing the shared resource unlike a thread that is holding a mutex. Among other things, the use of serializing tokens prevents many of the situations that could result in deadlocks and priority inversions when using mutexes, as well as greatly simplifying the design and implementation of a many-step procedure that would require a resource to be shared among multiple threads. The serializing token code is evolving into something quite similar to the " Read-copy-update" feature now available in Linux. Unlike Linux's current RCU implementation, DragonFly's is being implemented such that only processors competing for the same token are affected rather than all processors in the computer.

DragonFly switched to multiprocessor safe slab allocator, which requires neither mutexes nor blocking operations for memory assignment tasks. It was eventually ported into standard C library in the userland, where it replaced FreeBSD's malloc implementation.

Virtual kernel

Since release 1.8 DragonFly has a virtualization mechanism similar to User-mode Linux, allowing a user to run another kernel in the userland. The virtual kernel (''vkernel'') is run in completely isolated environment with emulated network and storage interfaces, thus simplifying testing kernel subsystems and clustering features.

The vkernel has two important differences from the real kernel: it lacks many routines for dealing with the low-level hardware management and it uses C standard library (libc) functions in place of in-kernel implementations wherever possible. As both real and virtual kernel are compiled from the same code base, this effectively means that platform-dependent routines and re-implementations of libc functions are clearly separated in a source tree.

The vkernel runs on top of hardware abstractions provided by the real kernel. These include the kqueue-based timer, the console (mapped to the virtual terminal where vkernel is executed), the disk image and virtual kernel Ethernet device (''VKE''), tunneling all packets to the host's tap interface.

Package management

Third-party software is available on DragonFly as binary packages via pkgng or from a native ports collection – DPorts.

DragonFly originally used the FreeBSD Ports collection as its official package management system, but starting with the 1.4 release switched to NetBSD's pkgsrc system, which was perceived as a way of lessening the amount of work needed for third-party software availability. Eventually, maintaining compatibility with pkgsrc proved to require more effort than was initially anticipated, so the project created DPorts, an overlay on top of the FreeBSD Ports