Great info guys,  I'll forward it on to him.

Danny


On Thu, 23 Sep 2004 16:41:06 -0600, Josh Coates <[EMAIL PROTECTED]> wrote:
> it's been a while, but when i used to work with sparcs, cc was much better
> than gcc if were trying to really tweak things out.
> 
> i pulled up some old flags to crank out some great mflops on the ultra sparc
> chip:
> -xO5 -xchip=ultra -xarch=v8plusa
> 
> it was solaris 2.6 on an ultra-1
> 
> you may want to google around and see what's new in the world of solaris
> compiler optimizations.
> 
> and of course, your dual processor machine wont help you much if this is a
> single process application.
> 
> Josh Coates
> http://www.jcoates.org
> 
> 
> 
> -----Original Message-----
> From: [EMAIL PROTECTED]
> [mailto:[EMAIL PROTECTED] Behalf Of Daniel Smith
> Sent: Thursday, September 23, 2004 4:14 PM
> To: BYU Unix Users Group
> Subject: [uug] Compiling for Multi-Processor Machine
> 
> A coworker had the follwoing question.  I was wondering if anyone out
> there knew what need to be done.
> 
> "We have a simulator that is being targetted for a Sun Fire V440
> server with two 1.065 MHz processors and 4 GB memory.
> When we build the application then compare its execution times on the
> server with the exec time on a 550MHz 1GB workstation, the app runs
> faster on the workstation (by about 40%!).
> 
> My questions:
> 
> Are there specific compiler options that we need to identify to
> optimize for the multiprocessor environment?
> 
> We are currently running Solaris 9.0 and using the gcc compiler v
> 3.2.2.  Are there more suitable tools available (e.g. the Sun Compiler
> suite)?"
> 
> Any thoughts?
> 
> Danny
> 
> ____________________
> BYU Unix Users Group
> http://uug.byu.edu/
> ___________________________________________________________________
> List Info: http://uug.byu.edu/cgi-bin/mailman/listinfo/uug-list
> 
>

____________________
BYU Unix Users Group 
http://uug.byu.edu/ 
___________________________________________________________________
List Info: http://uug.byu.edu/cgi-bin/mailman/listinfo/uug-list

Reply via email to