XNU
The XNU kernel
DeveloperApple Inc.
Written inC, C++
OS familyUnix-like, Unix
Working stateCurrent
Source modelOpen-source
Initial releaseDecember 1996 (1996-12)
Repository
PlatformsCurrent: x86-64, 64-bit ARM, 32-bit ARM (32-bit ARM support is closed-source)
Historical: PowerPC (32-bit and 64-bit), IA-32
Kernel typeHybrid
LicenseApple Public Source License 2.0
Preceded byMach, Nukernel
Official websitegithub.com/apple-oss-distributions/xnu

XNU (X is Not Unix) is the computer operating system (OS) kernel developed at Apple Inc. since December 1996 for use in the Mac OS X (now macOS) operating system and released as free and open-source software as part of the Darwin OS, which in addition to macOS is also the basis for the Apple TV Software, iOS, iPadOS, watchOS, visionOS, and tvOS OSes.[1]

XNU was originally developed by NeXT for the NeXTSTEP operating system. It was a hybrid kernel derived from version 2.5 of the Mach kernel developed at Carnegie Mellon University, which incorporated the bulk of the 4.3BSD kernel modified to run atop Mach primitives, along with an application programming interface (API) in Objective-C for writing drivers named Driver Kit.[2]

After Apple acquired NeXT, the kernel was updated with code derived from OSFMK 7.3 from OSF,[3] and the FreeBSD project, and the Driver Kit was replaced with new API on a restricted subset of C++[4] (based on Embedded C++)[5] named I/O Kit.

Kernel design

XNU is a hybrid kernel, containing features of both monolithic kernels and microkernels, attempting to make the best use of both technologies, such as the message passing ability of microkernels enabling greater modularity and larger portions of the OS to benefit from memory protection, and retaining the speed of monolithic kernels for some critical tasks.

As of 2021, XNU runs on ARM64 and x86-64 processors, both one processor and symmetric multiprocessing (SMP) models. PowerPC support was removed as of the version in Mac OS X Snow Leopard. Support for IA-32 was removed as of the version in Mac OS X Lion; support for 32-bit ARM was removed as of the version in iOS 11.

Mach

The basis of the XNU kernel is a heavily modified (hybrid) Open Software Foundation Mach kernel (OSFMK) 7.3.[3] OSFMK 7.3 is a microkernel[6] that includes applicable code from the University of Utah Mach 4 kernel and from the many Mach 3.0 variants forked from the original Carnegie Mellon University Mach 3.0 microkernel.

OSFMK 7.3 is able to run the core of an operating system as separated processes, which allows a great flexibility (it could run several operating systems in parallel above the Mach core), but this often reduces performance because of time-consuming kernel/user mode context switches and overhead stemming from mapping or copying messages between the address spaces of the kernel and that of the service daemons.

Apple licensed OSFMK 7.3 from the OSF, and attempted to streamline some tasks by building BSD functions into the kernel along with the Mach code. The result is a heavily modified (hybrid) OSFMK 7.3 kernel.

BSD

The Berkeley Software Distribution (BSD) part of the kernel provides the Portable Operating System Interface (POSIX) application programming interface (API, BSD system calls), the Unix process model atop Mach tasks, basic security policies, user and group ids, permissions, the network protocol stack (protocols), the virtual file system code (including a file system independent journaling layer), several local file systems such as Hierarchical File System (HFS, HFS Plus (HFS+)) and Apple File System (APFS), the Network File System (NFS) client and server, cryptographic framework, UNIX System V inter-process communication (IPC), audit subsystem, mandatory access control, and some of the locking primitives.[7] The BSD code present in XNU has been most recently synchronised with that from the FreeBSD kernel. Although much of it has been significantly modified, code sharing still occurs between Apple and the FreeBSD Project as of 2009.[8]

K32/K64

XNU in Mac OS X Snow Leopard, v10.6, (Darwin version 10) comes in two varieties, a 32-bit version called K32 and a 64-bit version called K64.[9] K32 can run 64-bit applications in userland.[10] What was new in Mac OS X 10.6 was the ability to run XNU in 64-bit kernel space. K32 was the default kernel for 10.6 Server when used on all machines except Mac Pro and Xserve models from 2008 onwards[11] and can run 64-bit applications. K64 has several benefits compared to K32:[12]

  • Can manage more than 32 GB RAM, as the memory map would consume a disproportionately large area of the 32-bit kernel space.
  • Cache buffer sizes can be larger than what the 32-bit kernel space allows, potentially increasing I/O performance.
  • Performance is increased when using high-performance networking devices or multiple graphics processing units (GPUs), as the kernel can map all of the devices in 64-bit space even if several have very large direct memory access (DMA) buffers.

Booting while holding down 6 and 4 forces the machine to boot K64 on machines supporting 64-bit kernels.[13] K64 will run 32-bit applications but it will not run 32-bit kernel extensions (KEXTs), so these must be ported to K64 to be able to load.

XNU in OS X Mountain Lion (10.8) and later only provides a 64-bit kernel.

I/O Kit

I/O Kit is the device driver framework, written in a subset of C++ based on Embedded C++.[14] Using its object-oriented design, features common to any class of driver are provided within the framework, helping device drivers be written in less time and code. The I/O Kit is multi-threaded, symmetric multiprocessing (SMP)-safe, and allows for hot-pluggable devices and automatic, dynamic device configuration.

Many drivers can be written to run from user space, which further enhances the stability of the system. If a user-space driver crashes, it will not crash the kernel. However, if a kernel-space driver crashes it will crash the kernel. Examples of kernel-space drivers include disk adapter and network adapter drivers, graphics drivers, drivers for Universal Serial Bus (USB) and FireWire host controllers, and drivers for virtual machine software such as VirtualBox, Parallels Desktop for Mac, and VMware Fusion.

See also

References

  1. ↑ "Porting UNIX/Linux Applications to Mac OS X: Glossary". Apple Computer. 2005. Retrieved June 7, 2017.
  2. ↑ "XNU GitHub source code README". GitHub. December 16, 2021.
  3. 1 2 Magee, Jim. WWDC 2000 Session 106 – Mac OS X: Kernel. 14 minutes in. Archived from the original on December 21, 2021.
  4. ↑ "The libkern C++ Runtime". IOKit Device Driver Design Guidelines.
  5. ↑ Amit Singh. "What is Mac OS X?". Archived from the original on April 19, 2019. Retrieved March 4, 2015.
  6. ↑ Wells, Douglas M. (1994). A Trusted, Scalable, Real-Time Operating System Environment (PDF). 1994 IEEE Dual-Use Technologies and Applications Conference. S2CID 5205380. Archived from the original (PDF) on August 22, 2017.
  7. ↑ Watson, Robert (August 2, 2008). "Re: freebsd-advocacy Digest, Vol 248, Issue 1". freebsd-advocacy (Mailing list). Retrieved October 24, 2013.
  8. ↑ "FreeBSD Quarterly Status Report (Grand Central Dispatch – FreeBSD port)". April 2009. Retrieved October 24, 2013.
  9. ↑ Mac OS X 10.6 Snow Leopard: the Ars Technica review, page 5
  10. ↑ Siracusa, John (August 31, 2009). "Mac OS X 10.6 Snow Leopard: the Ars Technica review". Ars Technica. Finally, this is worth repeating: please keep in mind that you do not need to run the 64-bit kernel in order to run 64-bit applications or install more than 4 GB of RAM in your Mac. Applications run just fine in 64-bit mode on top of the 32-bit kernel, and even in earlier versions of Mac OS X it's been possible to install and take advantage of much more than 4GB of RAM.
  11. ↑ Mac OS X Server v10.6: Macs that use the 64-bit kernel
  12. ↑ "What's New in Mac OS X: Mac OS X v10.6". Apple Developer. Apple, Inc.
  13. ↑ Mac OS X Server v10.6: Starting up with the 32-bit or 64-bit kernel
  14. ↑ Amit Singh. "XNU: The Kernel". What is Mac OS X?. I/O Kit. Archived from the original on April 19, 2019. Retrieved September 9, 2011. I/O Kit uses a restricted subset of C++
This article is issued from Wikipedia. The text is licensed under Creative Commons - Attribution - Sharealike. Additional terms may apply for the media files.