User: Password:
|
|
Subscribe / Log in / New account

Object-oriented design patterns in the kernel, part 2

Object-oriented design patterns in the kernel, part 2

Posted Jun 8, 2011 11:10 UTC (Wed) by cesarb (subscriber, #6266)
In reply to: Object-oriented design patterns in the kernel, part 2 by jengelh
Parent article: Object-oriented design patterns in the kernel, part 2

The Linux kernel runs on two types of processors: 32-bit processors where both pointers and unsigned long are 32-bit, and 64-bit processors where both pointers and unsigned long are 64-bit.

I do not think the kernel developers will bother until they decide to port to a processor where pointers are 128-bit.

For userspace programs, it is different since at least one unnamed operating system has 64-bit pointers but only 32-bit unsigned long when running on a 64-bit processor.


(Log in to post comments)

Object-oriented design patterns in the kernel, part 2

Posted Jun 8, 2011 20:00 UTC (Wed) by arnd (subscriber, #8866) [Link]

There was recently an attempt to port the kernel to a 48 bit pointer architecture that uses 16 bit segment registers with 32 bit offsets. The replies from kernel developers towards this were between "no way" and "as long as you fix all the bugs and make the code look prettier in the process."

Object-oriented design patterns in the kernel, part 2

Posted Jun 12, 2011 17:33 UTC (Sun) by josh (subscriber, #17465) [Link]

Which architecture has that fascinating addressing scheme?

Object-oriented design patterns in the kernel, part 2

Posted Jun 16, 2011 16:01 UTC (Thu) by siride (guest, #62756) [Link]

i386 of course


Copyright © 2017, Eklektix, Inc.
Comments and public postings are copyrighted by their creators.
Linux is a registered trademark of Linus Torvalds