(no subject)

[Dennis Castleman]

[-- Attachment #1: Type: text/plain, Size: 132 bytes --]

ALL

Anybody know the performance differences I can expect using a MIPS 5K core
@250 Mhz in 64bit mode versus 32bit mode?

Dennis 


[-- Attachment #2: Type: text/html, Size: 539 bytes --]

Re: your mail

[Jun Sun]

On Thu, Apr 17, 2003 at 10:53:57AM -0700, Dennis Castleman wrote:
> ALL
> 
> Anybody know the performance differences I can expect using a MIPS 5K core
> @250 Mhz in 64bit mode versus 32bit mode?
>

It really depends on the applications.  The biggest gain from 64bit,
other than the obviously bigger address space, is 64bit data
manipulation.  A single 64bit instruction (add/sub/...) is carried
out by several instructions in 32bit.

If your apps are heavy on 64bit operations, then you gain.
Otherwise I suspect no performance gain or even possibly a little 
worse performance due more stress on cache/memory subsystems.

I am sure others with more 64bit experience probably have more 
to say.  :)


Jun

Re: your mail

[Greg Lindahl]

On Thu, Apr 17, 2003 at 11:17:10AM -0700, Jun Sun wrote:

> It really depends on the applications.  The biggest gain from 64bit,
> other than the obviously bigger address space, is 64bit data
> manipulation.  A single 64bit instruction (add/sub/...) is carried
> out by several instructions in 32bit.

A big gain is the increased # of registers and better calling
sequence. Everyone sees that, not just people who want to use 64-bit
integers. At the moment you need to run the 64-bit kernel -- and the
new binutils & glibc -- in order to get n32 programs to work.

-- greg

Re: your mail

[Ralf Baechle]

On Thu, Apr 17, 2003 at 10:53:57AM -0700, Dennis Castleman wrote:

> Anybody know the performance differences I can expect using a MIPS 5K core
> @250 Mhz in 64bit mode versus 32bit mode?

As a rule of thumb - less performance.  64-bit code is typically larger
resulting in lower cache hit rate.  And since performance optimization
these days is essentially equivalent to maximising the cache hit rate
going 64-bit usually means a performance drop due to the drastically
larger size of code and data.

On the positive side for 64-bit stuff there's the possibility to do
64-bit computations with just one instruction, move data with less
instructions, use twice as many double precission fp registers that are
offered by 64-bit ABIs and more calling sequences.

The first two paragraphs were sort of a generic statement regarding
32-bit vs. 64-bit software on MIPS processors and affect both kernel and
userspace.  There's a few additional issues with the Linux kernel.
The 32-bit kernels requires all memory to be addressable through KSEG0
which limits it to at most 512MB; typically the limit is more like 256MB.
Memory above 512MB physical address can only be used as highmem.  That's
fairly inefficient and requires alot of special care when writing new
kernel software.  For processors that suffer from virtual aliases in
their data cache highmem currently is frighteningly inefficient - and
high memory pressure on lowmem doesn't help either.  So from a certain
point on that's simply making a 64-bit kernel is simply the better
idea - even for running 32-bit software.  That in particular applies
to very I/O intensive stuff.

In short - the right choice is a tradeoff between the hardware platform
and the application's requirements.  Choose wrong and you'll curse
loudly :-)

  Ralf