Re: 32-bit vs. 64-bit x86 Speed

"Michael Tiomkin" <tmk@netvision.net.il>
13 Apr 2007 01:37:26 -0400

          From comp.compilers

Related articles
[2 earlier articles]
Re: 32-bit vs. 64-bit x86 Speed marcov@stack.nl (Marco van de Voort) (2007-04-13)
Re: 32-bit vs. 64-bit x86 Speed anton@mips.complang.tuwien.ac.at (2007-04-13)
Re: 32-bit vs. 64-bit x86 Speed DrDiettrich1@aol.com (Hans-Peter Diettrich) (2007-04-13)
Re: 32-bit vs. 64-bit x86 Speed ian.rogers@manchester.ac.uk (Ian Rogers) (2007-04-13)
Re: 32-bit vs. 64-bit x86 Speed meissner@the-meissners.org (Michael Meissner) (2007-04-13)
Re: 32-bit vs. 64-bit x86 Speed georgeps@xmission.com (George Peter Staplin) (2007-04-13)
Re: 32-bit vs. 64-bit x86 Speed tmk@netvision.net.il (Michael Tiomkin) (2007-04-13)
Re: 32-bit vs. 64-bit x86 Speed dot@dotat.at (Tony Finch) (2007-04-13)
Re: 32-bit vs. 64-bit x86 Speed kenney@cix.compulink.co.uk (2007-04-13)
Re: 32-bit vs. 64-bit x86 Speed DrDiettrich1@aol.com (Hans-Peter Diettrich) (2007-04-14)
Re: 32-bit vs. 64-bit x86 Speed DrDiettrich1@aol.com (Hans-Peter Diettrich) (2007-04-14)
Re: 32-bit vs. 64-bit x86 Speed gah@ugcs.caltech.edu (glen herrmannsfeldt) (2007-04-18)
Re: 32-bit vs. 64-bit x86 Speed haberg@math.su.se (2007-04-23)
[5 later articles]
| List of all articles for this month |
From: "Michael Tiomkin" <tmk@netvision.net.il>
Newsgroups: comp.compilers
Date: 13 Apr 2007 01:37:26 -0400
Organization: Compilers Central
References: 07-04-031
Keywords: architecture, performance
Posted-Date: 13 Apr 2007 01:37:25 EDT

On Apr 12, 5:33 am, Jon Forrest <jlforr...@berkeley.edu> wrote:
> I don't think it's worth messing with 64-bit computing for apps that
> don't need the address space."


      Recall that "moving bits around" is done in parallel, and filling a
64-bit register usually takes the same time as filling a 16- or 32-
bit reg. Some operations can easily benefit from wider registers,
e.g. large string compare, memory move, bit operations on large
fields, large integer arithmetic (encryption). This means that even if
you only do file I/O, TCP/IP and encryption, 64-bit ops may help.


> Let's say you're a Linux user who never needs to run programs that
> don't fit in 32-bits. Would you run a 32-bit or a 64-bit version of
> Linux? You compiler people probably have intimate knowledge of the ISA
> issues here so I'm interested in what you have to say.


      I remember enojoying running Windows 3.11, a 32-bit OS with 16-bit
user processes, much better than the previous versions of Windows!-)
At that time people also said that most users do not need 32-bit apps.
      BTW, what does it mean "never needs to run programs that don't fit
in 32-bits"? I remember somebody saying that a home computer will
never need more than 64KB memory.


      Another question is whether your user needs better performance. Any
mulimedia creation application will happily eat all your CPU and
memory resources.


And of course you might be right, and meanwhile 64-bit Linux apps can
be less efficient - the data can take more memory because different
alignment, or they can define 'int' as a 32-bit int, and then most of
your programs wouldn't benefit from the 64-bit ints. Did you see the
benchmark comparison of 32-bit and 64-bit Linuxes/typical application
load?


      Michael



Post a followup to this message

Return to the comp.compilers page.
Search the comp.compilers archives again.