Re: fortran/3743: Reference to intrinsic `ISHFT' invalid

Re: fortran/3743: Reference to intrinsic `ISHFT' invalid

[Billinghurst, David (CRTS)]

The following reply was made to PR fortran/3743; it has been noted by GNATS.

From: "Billinghurst, David (CRTS)" <David.Billinghurst@riotinto.com>
To: <gcc-gnats@gcc.gnu.org>
Cc:  
Subject: Re: fortran/3743: Reference to intrinsic `ISHFT' invalid
Date: Wed, 23 Jan 2002 14:12:16 +0800

 -----Original Message-----
 From: David Edelsohn [mailto:dje@watson.ibm.com]
 Sent: Tuesday, 1 January 2002 7:09=20
 To: Billinghurst, David (CRTS); Toon Moene
 Cc: gcc-patches@gcc.gnu.org
 Subject: Re: Testcase for fortran/PR3743 "Reference to intrinsic `ISHFT'
 invalid".
 
 
 	I tracked down the underlying cause of the failure, but the cause
 is a fundamental design flaw in g77.
 
 	The symptom is intrin.c:ffeintrin_is_intrinsic() returns FALSE for
 some Fortran intrinsics.  Intrinsics are checked using bsearch() on the
 array of intrinsics in intrin.def.  The names are compared in the case
 determined by ffe_case_intrin() which is initialized to=20
 FFETARGET_defaultCASE_INTRIN in top.c, using the default value
 FFE_caseLOWER in target.h.
 
 	bsearch() requires a sorted array.  The failing testcase,
 BIT_SIZE, includes an underscore in the name.  Underscore is collated
 between uppercase letter and lowercase letters.  intrin.def is sorted =
 for
 uppercase names.
 
 DEFNAME ("BITEST",      "bitest",       "BITest",       genNONE, =
 specBITEST)     /* VXT */
 DEFNAME ("BIT_SIZE",    "bit_size",     "Bit_Size",     genNONE, =
 specBIT_SIZE)   /* F90 */
 DEFNAME ("BJTEST",      "bjtest",       "BJTest",       genNONE, =
 specBJTEST)     /* VXT */
 
 bsearch() looks through the array using a lowercase comparison and =
 expects
 to find bit_size collated before bitest, not BIT_SIZE collated after
 BITEST.  This problem may be the cause of other errors with respect to
 types of intrinsics.
 
 	Because of g77's options for cases, re-sorting the array or
 changing the default case would only cover up the problem.  libiberty
 provides strcasecmp, so maybe that should be used for the comparison
 instead of writing all case variants.
 
 David

Re: fortran/3743: Reference to intrinsic `ISHFT' invalid

[toon]

Synopsis: Reference to intrinsic `ISHFT' invalid

Responsible-Changed-From-To: unassigned->toon
Responsible-Changed-By: toon
Responsible-Changed-When: Thu Jan 31 11:34:23 2002
Responsible-Changed-Why:
    My job.
State-Changed-From-To: analyzed->closed
State-Changed-By: toon
State-Changed-When: Thu Jan 31 11:34:23 2002
State-Changed-Why:
    Fixed in 3.1

http://gcc.gnu.org/cgi-bin/gnatsweb.pl?cmd=view%20audit-trail&database=gcc&pr=3743

Re: fortran/3743: Reference to intrinsic `ISHFT' invalid

[Billinghurst, David (CRTS)]

The following reply was made to PR fortran/3743; it has been noted by GNATS.

From: "Billinghurst, David (CRTS)" <David.Billinghurst@riotinto.com>
To: <gcc-gnats@gcc.gnu.org>
Cc:  
Subject: Re: fortran/3743: Reference to intrinsic `ISHFT' invalid
Date: Tue, 15 Jan 2002 12:02:35 +0800

 Further analysis of this PR, and a suggested fix, at:
 
 http://gcc.gnu.org/ml/gcc-patches/2002-01/msg00000.html
 http://gcc.gnu.org/ml/gcc-patches/2002-01/msg00049.html
 http://gcc.gnu.org/ml/gcc-patches/2002-01/msg00091.html
 
 

Re: fortran/3743: Reference to intrinsic `ISHFT' invalid

[dje]

Synopsis: Reference to intrinsic `ISHFT' invalid

State-Changed-From-To: open->analyzed
State-Changed-By: dje
State-Changed-When: Mon Dec 31 15:15:18 2001
State-Changed-Why:
    G77 does not recognize BIT_SIZE intrinsic as integer.  The
    implicit type for a function whose name begins with the
    letter 'B' is REAL, so G77 reports that a function returning
    basictypeREAL is not a valid argument to the ISHFT intrinsic
    expecting a basictypeINTEGER argument.  Declaring BIT_SIZE
    as INTEGER removes the error.
    It is unclear why this only fails on some targets.  BIT_SIZE
    does not seem to be recognized as an intrinsic when the
    arguments to ISHFT are examined, it considers BIT_SIZE a
    function returning REAL.  Is something non-deterministic
    in the parser?

http://gcc.gnu.org/cgi-bin/gnatsweb.pl?cmd=view%20audit-trail&database=gcc&pr=3743

RE: fortran/3743: Reference to intrinsic `ISHFT' invalid

[Billinghurst, David (CRTS)]

The following reply was made to PR fortran/3743; it has been noted by GNATS.

From: "Billinghurst, David (CRTS)" <David.Billinghurst@riotinto.com>
To: "'gcc-gnats@gcc.gnu.org'" <gcc-gnats@gcc.gnu.org>, nobody@gcc.gnu.org
Cc: gcc-prs@gcc.gnu.org
Subject: RE: fortran/3743: Reference to intrinsic `ISHFT' invalid
Date: Fri, 27 Jul 2001 00:28:34 -0000

 This may be related to PR fortran/3807: Function BESJN(integer,double)
 problems
 
 

fortran/3743: Reference to intrinsic `ISHFT' invalid

[David.Billinghurst]

>Number:         3743
>Category:       fortran
>Synopsis:       Reference to intrinsic `ISHFT' invalid
>Confidential:   no
>Severity:       serious
>Priority:       medium
>Responsible:    unassigned
>State:          open
>Class:          sw-bug
>Submitter-Id:   net
>Arrival-Date:   Thu Jul 19 23:06:06 PDT 2001
>Closed-Date:
>Last-Modified:
>Originator:     David Billinghurst
>Release:        2.95.2 and 3.1 CVS
>Organization:
>Environment:
mips-sgi-irix6.5
>Description:
Attached program gives error on mips-sgi-irix6.5, but not on i686-pc-cygwin or SGI f90

Error is:
j.f: In program `MAIN__':
j.f:14: 
         i3 = ISHFT(i,BIT_SIZE(i))
              ^
Reference to intrinsic `ISHFT' at (^) invalid -- one or more arguments have incorrect type


According to F90 standard (section 15.13)
* the first arg of ISHIFT is integer type
* the second arg of ISHIFT is integer type
* the return type of BIT_SIZE is same as argument
* the return type of ISHFT is same as the first arg
 

####################################

c On mips-sgi-irix6.5 get
c
c         i3 = ISHFT(i,BIT_SIZE(i))
c              ^
c Reference to intrinsic `ISHFT' at (^) invalid -- 
c one or more arguments have incorrect type
c
c     David Billinghurst <David.Billinghurst@riotinto.com>
c
      integer   i, i2, i3
      i  = 3
      i2 = BIT_SIZE(i)
      i3 = ISHFT(i,i2)
      i3 = ISHFT(i,BIT_SIZE(i))
      end
>How-To-Repeat:

>Fix:

>Release-Note:
>Audit-Trail:
>Unformatted: