This is the mail archive of the gcc-help@gcc.gnu.org mailing list for the GCC project.

Index Nav:	[Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav:	[Date Prev] [Date Next]	[Thread Prev] [Thread Next]
Other format:	[Raw text]

Re: gfortran 4.2 with openMP: why no speedup?

From: Tim Prince <tprince at myrealbox dot com>
To: Anand Patil <anand dot prabhakar dot patil at gmail dot com>
Cc: "Nelson H. F. Beebe" <beebe at math dot utah dot edu>, gcc-help at gcc dot gnu dot org
Date: Sat, 12 Jan 2008 14:44:50 -0800
Subject: Re: gfortran 4.2 with openMP: why no speedup?
References: <2bc7a5a50801111718o6fb77b21yfd83638afa02e313@mail.gmail.com> <CMM.0.95.0.1200104348.beebe@psi.math.utah.edu> <2bc7a5a50801121434u4cbd8bacx545aad7e281ff624@mail.gmail.com>

Anand Patil wrote:
attribution dropped:
>> You should also check the generated assembly code (f77 -S foo.f)
>> whether C(i,j)**2 is compiled into the inline code C(i,j)*C(i,j), or
>> into call to the run-time library power function, and also whether the
>> subscript address computations are eliminated.
> 
> I'll just inline it manually to be sure. I was trying to get a speedup
> from openMP in that subroutine, not necessarily optimize overall.
> 

It's true that when your only objective is to get speedup from
threading, a number of normal optimizations may make your task more
difficult.  I don't know of any current Fortran compilers which don't
optimize **2 (as opposed to **2.) as soon as any optimization is enabled.

References:
- gfortran 4.2 with openMP: why no speedup?
  - From: Anand Patil
- Re: gfortran 4.2 with openMP: why no speedup?
  - From: Anand Patil

Index Nav:	[Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav:	[Date Prev] [Date Next]	[Thread Prev] [Thread Next]