Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Path: utzoo!mnetor!seismo!mimsy!aplcen!jhunix!ecf_ujaj
From: ecf_ujaj@jhunix.UUCP (Jim Jones)
Newsgroups: comp.lang.fortran
Subject: IBM and Fortran bit-ops (was FORTRAN-77 XLA PLUS BITWISE OPERATORS)
Message-ID: <4788@jhunix.UUCP>
Date: Wed, 29-Apr-87 10:38:28 EDT
Article-I.D.: jhunix.4788
Posted: Wed Apr 29 10:38:28 1987
Date-Received: Sat, 2-May-87 11:10:47 EDT
References: <12296045502.17.BEEBE@SCIENCE.UTAH.EDU>
Reply-To: ecf_ujaj@jhunix.UUCP (Jim Jones)
Organization: Johns Hopkins Univ. Computing Ctr.
Lines: 24

<is this still necessary?>

In article <12296045502.17.BEEBE@SCIENCE.UTAH.EDU> Beebe@SCIENCE (Nelson H.F. Beebe) writes:
>...
>integer bit operations  (.AND., .OR.,  .NOT., .XOR.),  since
>almost all compilers (IBM notably excepted) have implemented
>them in some fashion.

 It is true that logical operators cannot be used on integer variables
 with IBM's current flagship Fortran compiler (VS-Fortran).  However,
 the run-time library supplied with the compiler does provide support
 for a number of bit oriented services.  In addition to AND, OR, NOT,
 and XOR, bit testing, setting and clearing are possible with INTEGER
 variables.  While this scenario is not as nice as simply allowing
 logical operators to act on INTEGER variables (subroutine calls are
 much less efficient than inline code!), we've not been left completely
 out in the cold this time...

 -jj
-- 

 UUCP:  {ihnp4!whuxcc | seismo!umcp-cs}!jhunix!ecf_ujaj
 ARPA:  jones@hopkins.arpa
 BITNET:  L64A0110@jhuvm