home *** CD-ROM | disk | FTP | other *** search
- Xref: sparky comp.sys.amiga.programmer:17528 comp.sys.amiga.hardware:21767
- Path: sparky!uunet!zaphod.mps.ohio-state.edu!moe.ksu.ksu.edu!crcnis1.unl.edu!cse.unl.edu!tbills
- From: tbills@cse.unl.edu (Trent Bills)
- Newsgroups: comp.sys.amiga.programmer,comp.sys.amiga.hardware
- Subject: Re: CISC and RISC
- Date: 18 Dec 1992 14:57:00 GMT
- Organization: University of Nebraska--Lincoln
- Lines: 36
- Distribution: world
- Message-ID: <1gsorsINN327@crcnis1.unl.edu>
- References: <70436@cup.portal.com> <amipb.04wr@amipb.gna.org> <37844@cbmvax.commodore.com> <AHANSFOR.92Dec17115522@bigwpi.wpi.edu>
- NNTP-Posting-Host: cse.unl.edu
-
- In article <AHANSFOR.92Dec17115522@bigwpi.wpi.edu>, ahansfor@bigwpi.wpi.edu (Andrew L. Hansford) writes:
-
- |> tbills> The extreme cost of task switching is part of the motivation
- |> tbills> for putting multiple processors in one machine. Most (all?)
- |> tbills> workstation makers have some kind of multiprocessor machine.
- |>
- |> tbills> |> -- Andrew Hansford |> ahansfor@wpi.wpi.edu
- |>
- |> tbills> - Trent Bills
- |>
- |> True. But like exception handling, more support for this seems to go
- |> into CISC machines. As for multiple processors, you still incur the
- |> same cost for a context switch. Now you have more processors
- |> switching. Esp with microkernel architectures that instead of one
- |> switch to the kernel and back to the calling function (2 switches),
- |> you get a context switch to the kernel, a message is copied or
- |> remapped, a context switch to a user lever server, and then the server
- |> enters the kernel to pass a message back to the calling proces. Now
- |> we have twice and many context switches.
-
- The idea behind having multiple processors is that you only have a few
- active processes at any one time and that each processor can be more
- or less dedicated to that process.
-
- The cost of a context switch in a RISC machine isn't usually that much
- more that a CISC machine. The only real difference is that the CISC
- machine only has to execute one or two instructions to perform the
- context switch whereas the RISC machine will have to execute much more.
- However, every instruction in the RISC machine will only take one clock
- cycle whereas the instructions for a task switch in a CISC machine
- will take a rather large number of clock cycles. Every access to memory
- will take at least one clock cycle regardless of the type of processor.
- The result is larger code on the RISC machine, but roughly identical
- performance for both.
-
- - Trent Bills
-