Re: Safety and legality of optimizations

clodius@lanl.gov (William B. Clodius)
Wed, 26 Apr 1995 18:34:41 GMT

          From comp.compilers
Related articles
[2 earlier articles]
Re: Safety and legality of optimizations Vinod.Grover@Eng.Sun.COM (1995-03-28)
Re: Safety and legality of optimizations cliffc@crocus.hpl.hp.com (1995-04-03)
Re: Safety and legality of optimizations cef@geodesic.com (Charles Fiterman) (1995-04-15)
Re: Safety and legality of optimizations Dave@OCCL-CAM.DEMON.CO.UK (Dave Lloyd) (1995-04-18)
Re: Safety and legality of optimizations cliffc@crocus.hpl.hp.com (1995-04-17)
Re: Safety and legality of optimizations jim@meiko.co.uk (1995-04-21)
Re: Safety and legality of optimizations clodius@lanl.gov (1995-04-26)
| List of all articles for this month |
Newsgroups: comp.compilers
From: clodius@lanl.gov (William B. Clodius)
Keywords: Fortran, optimize
Organization: NIS-1, Los Alamos National Laboratory
References: 95-02-179> 95-04-075
Date: Wed, 26 Apr 1995 18:34:41 GMT

Note that the next Fortran standard (F95) is expected to allow functions to
have a PURE attribute, for precisely the reasons discussed here. They are
also allowed to have an ELEMENTAL attribute, essentially a PURE function
with scalar arguments and result that is readilly vectorizable or
parallelizable. An ELEMENTAL function is also allowed to initialize
variables and PARAMETERs (constants) so that some user defined functions can
now be used for such initializations.


------------------------------------------------------------------------
William B. Clodius, wclodius@lanl.gov
NIS-1, Los Alamos National Laboratory
--

Post a followup to this message

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