The NetBSD Project is pleased to announce that NetBSD 5.0, the thirteenth major release of the NetBSD operating system, is now available. NetBSD 5.0 features greatly improved performance and scalability on modern multiprocessor (SMP) and multi-core systems. Multi-threaded applications can now efficiently make use of more than one CPU or core, and system performance is much better under I/O and network load, benefiting, for example, server, scientific, and software development workloads.

This improved performance is the result of a rewritten threading subsystem based on a 1:1 threading model, new kernel synchronization primitives, kernel preemption, a rewritten scheduler implementation, real-time scheduling extensions, processor sets, and dynamic CPU sets for thread affinity. Almost all core kernel subsystems, like virtual memory, memory allocators, file system frameworks for major file systems, and others were audited and overhauled to make use of highly concurrent algorithms.

Announcement | Changelog | Download

Known Problems

Using block device nodes directly for I/O may cause a kernel crash when the file system containing /dev is FFS and is mounted with -o log. Workaround: use raw disk devices, or remount the file system without -o log.

Occasionally, gdb may cause a process that is being debugged to hang when "single stepped". Workaround: kill and restart the affected process.

gdb cannot debug running threaded programs correctly. Workaround: generate a core file from the program using gcore(1) and pass the core to gdb, instead of debugging the running program.

Statically linked binaries using pthreads are currently broken.

Certain early revision AMD Opteron and Athlon 64 processors contain a bug that may cause system instability when running with more than one CPU core active. An OS-level workaround for this issue has been prepared but was not ready in time for inclusion in NetBSD 5.0. It will be available as part of a later release in the 5.0 series.