[Bug fortran/16579] New: gfortran: (ICHAR(CHAR(I)) .NE. I) when 128 <= I <= 255

[Bug fortran/16579] New: gfortran: (ICHAR(CHAR(I)) .NE. I) when 128 <= I <= 255

[billingd at gcc dot gnu dot org]

Test g77.f-torture/execute/20010610.f fails with gfortran.
Looks like a signed vs unsigned char issue.

The test is:

      DO I = 0, 255
         IF (ICHAR(CHAR(I)) .NE. I) CALL ABORT
      ENDDO
      END

It fails for I >= 128

ICHAR(CHAR(128)) = -128
ICHAR(CHAR(129)) = -127
 ...
ICHAR(CHAR(255)) = -1

-- 
           Summary: gfortran: (ICHAR(CHAR(I)) .NE. I)  when 128 <= I <= 255
           Product: gcc
           Version: 3.5.0
            Status: UNCONFIRMED
          Severity: normal
          Priority: P2
         Component: fortran
        AssignedTo: unassigned at gcc dot gnu dot org
        ReportedBy: billingd at gcc dot gnu dot org
                CC: gcc-bugs at gcc dot gnu dot org
  GCC host triplet: i686-pc-cygwin


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=16579

[Bug fortran/16579] gfortran: (ICHAR(CHAR(I)) .NE. I) when 128 <= I <= 255

[sgk at troutmask dot apl dot washington dot edu]

------- Additional Comments From sgk at troutmask dot apl dot washington dot edu  2004-07-16 04:24 -------

The standard says that the range for I in CHAR(I) "shall be of type integer
with a value in the range [0, n - 1], where n is the number of characters
in the collating sequence associated with the specified kind type parameter."
Gfortran uses the ASCII collating sequence, which has 128 characters.  The
standard does not specify the return value of CHAR(I) when I is not in the
specified range.  So, this isn't a signed versus an unsigned problem.  It
falls under the heading "processor defined behavior".  Personally, I think
CHAR() and ICHAR() should do the appropriate range check and issues an
error for out of range values.


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=16579

Re: [Bug fortran/16579] gfortran: (ICHAR(CHAR(I)) .NE. I) when 128 <= I <= 255

[Toon Moene]

sgk at troutmask dot apl dot washington dot edu wrote:

> ------- Additional Comments From sgk at troutmask dot apl dot washington dot edu  2004-07-16 04:24 -------
> 
> The standard says that the range for I in CHAR(I) "shall be of type integer
> with a value in the range [0, n - 1], where n is the number of characters
> in the collating sequence associated with the specified kind type parameter."
> Gfortran uses the ASCII collating sequence, which has 128 characters.  The
> standard does not specify the return value of CHAR(I) when I is not in the
> specified range.  So, this isn't a signed versus an unsigned problem.  It
> falls under the heading "processor defined behavior".  Personally, I think
> CHAR() and ICHAR() should do the appropriate range check and issues an
> error for out of range values.

I think it would be more useful to just allow the range 0..255 for the 
return value of ICHAR.  Too many codes depend on it.

I fixed this in 3.1 or 3.2 for g77 (when it suddenly went to -128 .. 127.

Cheers,

-- 
Toon Moene - mailto:toon@moene.indiv.nluug.nl - phoneto: +31 346 214290
Saturnushof 14, 3738 XG  Maartensdijk, The Netherlands
Maintainer, GNU Fortran 77: http://gcc.gnu.org/onlinedocs/g77_news.html
A maintainer of GNU Fortran 95: http://gcc.gnu.org/fortran/

[Bug fortran/16579] gfortran: (ICHAR(CHAR(I)) .NE. I) when 128 <= I <= 255

[toon at moene dot indiv dot nluug dot nl]

------- Additional Comments From toon at moene dot indiv dot nluug dot nl  2004-07-16 20:54 -------
Subject: Re:  gfortran: (ICHAR(CHAR(I)) .NE. I)  when 128
 <= I <= 255

sgk at troutmask dot apl dot washington dot edu wrote:

> ------- Additional Comments From sgk at troutmask dot apl dot washington dot edu  2004-07-16 04:24 -------
> 
> The standard says that the range for I in CHAR(I) "shall be of type integer
> with a value in the range [0, n - 1], where n is the number of characters
> in the collating sequence associated with the specified kind type parameter."
> Gfortran uses the ASCII collating sequence, which has 128 characters.  The
> standard does not specify the return value of CHAR(I) when I is not in the
> specified range.  So, this isn't a signed versus an unsigned problem.  It
> falls under the heading "processor defined behavior".  Personally, I think
> CHAR() and ICHAR() should do the appropriate range check and issues an
> error for out of range values.

I think it would be more useful to just allow the range 0..255 for the 
return value of ICHAR.  Too many codes depend on it.

I fixed this in 3.1 or 3.2 for g77 (when it suddenly went to -128 .. 127.

Cheers,



-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=16579

[Bug fortran/16579] gfortran: (ICHAR(CHAR(I)) .NE. I) when 128 <= I <= 255

[pinskia at gcc dot gnu dot org]

------- Additional Comments From pinskia at gcc dot gnu dot org  2004-07-17 03:18 -------
Confirmed.

-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|UNCONFIRMED                 |NEW
     Ever Confirmed|                            |1
           Keywords|                            |wrong-code
   Last reconfirmed|0000-00-00 00:00:00         |2004-07-17 03:18:53
               date|                            |


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=16579

[Bug fortran/16579] gfortran: (ICHAR(CHAR(I)) .NE. I) when 128 <= I <= 255

[sgk at troutmask dot apl dot washington dot edu]

------- Additional Comments From sgk at troutmask dot apl dot washington dot edu  2004-07-17 05:04 -------
I went looking for the bug and have a lead.  In simplify.c we 
have hard code numbers of 0 and 255 for CHAR() while we use
CHAR_MIN and CHAR_MAX from /usr/include/limits.h in ICHAR().
On FreeBSD (and suspect other systems), CHAR_MIN and CHAR_MAX
range from -128 to 127.  I haven't determined the correct fix
yet, but I believe we need to change line 1329--1331 from

  index = (int) e->value.character.string[0];

  if (index < CHAR_MIN || index > CHAR_MAX)

to 

 index = (int) ((unsigned int) e->value.character.string[0]);

  if (index < 0 || index > 255)

Although I hate hard coded numbers.



-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=16579

[Bug fortran/16579] gfortran: (ICHAR(CHAR(I)) .NE. I) when 128 <= I <= 255

[tobi at gcc dot gnu dot org]

------- Additional Comments From tobi at gcc dot gnu dot org  2004-07-21 21:15 -------
Also look at the beginning of simplify.c, ascii_table and xascii_table. These
are used for the frontend implemntations of char and ichar.

-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=16579

[Bug fortran/16579] gfortran: (ICHAR(CHAR(I)) .NE. I) when 128 <= I <= 255

[tobi at gcc dot gnu dot org]

------- Additional Comments From tobi at gcc dot gnu dot org  2004-08-29 19:16 -------
Patch here: http://gcc.gnu.org/ml/fortran/2004-08/msg00285.html
Lesson learned: actually looking at what happens can make writing patches easy :-)

-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
           Keywords|                            |patch


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=16579

[Bug fortran/16579] gfortran: (ICHAR(CHAR(I)) .NE. I) when 128 <= I <= 255

[cvs-commit at gcc dot gnu dot org]

------- Additional Comments From cvs-commit at gcc dot gnu dot org  2004-08-31 21:54 -------
Subject: Bug 16579

CVSROOT:	/cvs/gcc
Module name:	gcc
Changes by:	tobi@gcc.gnu.org	2004-08-31 21:54:25

Modified files:
	gcc/testsuite  : ChangeLog 
	gcc/fortran    : ChangeLog trans-types.c 
Added files:
	gcc/testsuite/gfortran.fortran-torture/execute: 
	                                                intrinsic_i_char.f90 

Log message:
	fortran/
	PR fortran/16579
	* trans-types.c (gfc_init_types): Make gfc_character1_type_node an
	unsigned char.
	
	testsuite/
	PR fortran/16579
	* gfortran.fortran-torture/execute/intrinsic_i_char.f90: New test.

Patches:
http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/testsuite/ChangeLog.diff?cvsroot=gcc&r1=1.4219&r2=1.4220
http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/testsuite/gfortran.fortran-torture/execute/intrinsic_i_char.f90.diff?cvsroot=gcc&r1=NONE&r2=1.1
http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/fortran/ChangeLog.diff?cvsroot=gcc&r1=1.194&r2=1.195
http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/fortran/trans-types.c.diff?cvsroot=gcc&r1=1.24&r2=1.25



-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=16579

[Bug fortran/16579] gfortran: (ICHAR(CHAR(I)) .NE. I) when 128 <= I <= 255

[billingd at gcc dot gnu dot org]

------- Additional Comments From billingd at gcc dot gnu dot org  2004-09-01 06:29 -------
Confirm fixed

-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|                            |FIXED


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=16579

[Bug fortran/16579] gfortran: (ICHAR(CHAR(I)) .NE. I) when 128 <= I <= 255

[cvs-commit at gcc dot gnu dot org]

------- Additional Comments From cvs-commit at gcc dot gnu dot org  2004-09-01 06:30 -------
Subject: Bug 16579

CVSROOT:	/cvs/gcc
Module name:	gcc
Changes by:	billingd@gcc.gnu.org	2004-09-01 06:30:06

Modified files:
	gcc/testsuite/gfortran.dg/g77: README 
	gcc/testsuite  : ChangeLog 
Added files:
	gcc/testsuite/gfortran.dg/g77: 20010610.f 

Log message:
	2004-09-01  David Billinghurst <David.Billinghurst@riotinto.com>
	
	PR fortran/16579
	* gfortran.dg/g77/20010610.f: Now passes.
	Copy from g77.f-torture/execute. Add dg-run directive.
	* gfortran.dg/g77/README: Update

Patches:
http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/testsuite/gfortran.dg/g77/20010610.f.diff?cvsroot=gcc&r1=NONE&r2=1.1
http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/testsuite/gfortran.dg/g77/README.diff?cvsroot=gcc&r1=1.5&r2=1.6
http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/testsuite/ChangeLog.diff?cvsroot=gcc&r1=1.4224&r2=1.4225



-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=16579

[Bug fortran/16579] gfortran: (ICHAR(CHAR(I)) .NE. I) when 128 <= I <= 255

[tobi at gcc dot gnu dot org]

------- Additional Comments From tobi at gcc dot gnu dot org  2004-09-01 21:12 -------
David, the testcase you checked in is identical to the one I checked in before
as gfortran.fortran-torture/execute/intrinsic_i_char.f90. I leave it up to you
to decide which one is superfluous and remove it.

I'm wondering if it wouldn't be better if we had the failing testcases from g77
as XFAILs in our testsuite, this way I wouldn't have missed that this PR
resulted from a g77 testcase.

-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |tobi at gcc dot gnu dot org


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=16579

[Bug fortran/16579] gfortran: (ICHAR(CHAR(I)) .NE. I) when 128 <= I <= 255

[david dot billinghurst at comalco dot riotinto dot com dot au]

------- Additional Comments From david dot billinghurst at comalco dot riotinto dot com dot au  2004-09-02 02:02 -------
Subject: RE:  gfortran: (ICHAR(CHAR(I)) .NE. I)  when 128 <= I <= 255

tobi at gcc dot gnu dot org wrote:
> David, the testcase you checked in is identical to the one I checked in before 
> as gfortran.fortran-torture/execute/intrinsic_i_char.f90. 

OK.  I will delete intrinsic_i_char.f90

> I'm wondering if it wouldn't be better if we had the failing testcases from g77 
> as XFAILs in our testsuite.

That is/was the plan.  I think I got all the easy ones.  Progress is recorded
in the gcc/testsuite/gfortran.dg/g77/README.  In this case, there was a chance
that the test was bogus.

__________________________________________________________________

NOTICE
 
This e-mail and any attachments are private and confidential and 
may contain privileged information.
 
If you are not an authorised recipient, the copying or distribution 
of this e-mail and any attachments is prohibited and you must not 
read, print or act in reliance on this e-mail or attachments.
 
This notice should not be removed.
__________________________________________________________________


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=16579

[Bug fortran/16579] gfortran: (ICHAR(CHAR(I)) .NE. I) when 128 <= I <= 255

[cvs-commit at gcc dot gnu dot org]

------- Additional Comments From cvs-commit at gcc dot gnu dot org  2004-09-03 03:29 -------
Subject: Bug 16579

CVSROOT:	/cvs/gcc
Module name:	gcc
Changes by:	billingd@gcc.gnu.org	2004-09-03 03:29:12

Modified files:
	gcc/testsuite  : ChangeLog 
Removed files:
	gcc/testsuite/gfortran.fortran-torture/execute: 
	                                                intrinsic_i_char.f90 

Log message:
	2004-09-03  David Billinghurst <David.Billinghurst@riotinto.com>
	
	PR fortran/16579
	* gfortran.fortran-torture/execute/intrinsic_i_char.f90:
	Delete.  Duplicate of gfortran.dg/g77/20010610.f

Patches:
http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/testsuite/ChangeLog.diff?cvsroot=gcc&r1=1.4234&r2=1.4235
http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/testsuite/gfortran.fortran-torture/execute/intrinsic_i_char.f90.diff?cvsroot=gcc&r1=1.1&r2=NONE



-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=16579

[Bug fortran/16579] gfortran: (ICHAR(CHAR(I)) .NE. I) when 128 <= I <= 255

[pinskia at gcc dot gnu dot org]

-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
   Target Milestone|---                         |4.0.0


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=16579