Re: vectorization in icc

kf@iki.fi
11 Dec 2002 22:24:55 -0500

          From comp.compilers

Related articles
[2 earlier articles]
vectorization in icc aart.bik@intel.com (Bik, Aart) (2002-12-03)
Re: vectorization in icc kfredrik@saippua.cs.Helsinki.FI (Kimmo Fredriksson) (2002-12-07)
vectorization in icc aart.bik@intel.com (Bik, Aart) (2002-12-07)
Re: vectorization in icc terryg@qwest.net (Terry Greyzck) (2002-12-11)
Re: vectorization in icc kf@iki.fi (2002-12-11)
Re: vectorization in icc kf@iki.fi (2002-12-11)
Re: vectorization in icc kf@iki.fi (2002-12-11)
Re: vectorization in icc nmm1@cus.cam.ac.uk (2002-12-13)
| List of all articles for this month |

From: kf@iki.fi
Newsgroups: comp.compilers
Date: 11 Dec 2002 22:24:55 -0500
Organization: -
References: 02-12-049
Keywords: parallel, performance
Posted-Date: 11 Dec 2002 22:24:55 EST

Hi again,


Things seem to change if you use larger values than 16 in the inner-loop
counter. Actually, the value is not hard coded, I just experimented with
16, because 128/8=16... In real world apps it can be anything. I tried
160, and the speed-up was about 7.2X, and about 2X over my hand coded
32bit vectorized code... So, things are looking better...


kf


Post a followup to this message

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