This is the mail archive of the fortran@gcc.gnu.org mailing list for the GNU Fortran 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: Loop Vectorization

From: Tobias Burnus <burnus at net-b dot de>
To: sdcycling <sdcycling at cox dot net>, tprince at computer dot org
Cc: fortran at gcc dot gnu dot org
Date: Mon, 20 Jun 2016 21:06:42 +0200
Subject: Re: Loop Vectorization
Authentication-results: sourceware.org; auth=none
References: <8E04A2EF-9140-4E84-81E7-61841073AF06 at cox dot net> <57678924 dot 7050807 at net-b dot de> <8vtq1t00B2qVqVd01vtrfL> <CFBD857D-4396-4660-B276-C12AC5735C25 at cox dot net>

sdcycling wrote:

I am currently using MPICH.   I could try MPICH in combination with OMP, but the number of processors on my workstation are currently maxed out.

Side note: OpenMP's "simd" directives work on vectorization and do notimply multiple threads. With -fopenmp-simd, the OpenMP run time libraryis not even linked.

Regarding the code the compiler produces, using "-fdump-tree-original",one can see what operations the compiler does. (Dump of the internalrepresentation, looks like low-level C, but not all properties areshown.) It's quite lengthy, but it might give a hint what code thecompiler adds.


Tobias

References:
- Loop Vectorization
  - From: sdcycling
- Re: Loop Vectorization
  - From: Tobias Burnus
- Re: Loop Vectorization
  - From: sdcycling

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