home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: comp.os.misc
- Path: sparky!uunet!zaphod.mps.ohio-state.edu!darwin.sura.net!mojo.eng.umd.edu!pandora.pix.com!stripes
- From: stripes@pix.com (Josh Osborne)
- Subject: Re: Processes and Translation Look-Aside Buffers
- Message-ID: <Bx99Eq.Cpp@pix.com>
- Sender: news@pix.com (The News Subsystem)
- Nntp-Posting-Host: pandora.pix.com
- Organization: Pix Technologies -- The company with no adult supervision
- References: <Bx7oDL.507@beach.csulb.edu>
- Date: Thu, 5 Nov 1992 18:16:00 GMT
- Lines: 26
-
- In article <Bx7oDL.507@beach.csulb.edu> werts@thrush.acs.csulb.edu (Michael Werts) writes:
- >Can anyone give me an idea of what is done with
- >the contents of a translation look-aside buffer
- >at task switch time in an operating system like
- >Unix with per process address spaces?
- >
- >Is the TLB reloaded? Or is it shared in someway
- >between processes?
-
- Most CPU/MMUs I have seen invalidate it. The AMD29k has a 16bit (it may
- be 8, but I think 16) task-id it uses during the look-up, so it can let
- TLBs live. I havn't seen any study showing when AMD's TLBs live long enough
- to make a diffrence, let alone a study showing the more genneral case (i.e.
- with a load of X and a task-id Y bits wide Z TLBs will live through enough
- context switchs to be re-used...).
-
- For cache there is a 3rd choice, you can use phyical addresses rather then
- vert. ones (at the cost of not being able to overlap cache look-up and
- VM maping), Intel does this with the 386 and 486. I don't know anyone
- else who does.
- --
- stripes@pix.com "Security for Unix is like
- Josh_Osborne@Real_World,The Multitasking for MS-DOS"
- "The dyslexic porgramer" - Kevin Lockwood
- We all agree on the necessity of compromise. We just can't agree on
- when it's necessary to compromise. - Larry Wall
-