[Bug fortran/21593] New: FAIL: gfortran.dg/dev_null.f90

[Bug fortran/21593] New: FAIL: gfortran.dg/dev_null.f90

[danglin at gcc dot gnu dot org]

Executing on host: /mnt/gnu/gcc-3.3/objdir/gcc/testsuite/../gfortran -B/mnt/gnu/
gcc-3.3/objdir/gcc/testsuite/../ /mnt/gnu/gcc-3.3/gcc/gcc/testsuite/gfortran.dg/
dev_null.f90   -O0   -pedantic-errors  -L/mnt/gnu/gcc-3.3/objdir/hppa2.0w-hp-hpu
x11.11/./libgfortran/.libs -L/mnt/gnu/gcc-3.3/objdir/hppa2.0w-hp-hpux11.11/./lib
iberty  -lm   -o ./dev_null.exe    (timeout = 300)
PASS: gfortran.dg/dev_null.f90  -O0  (test for excess errors)
Setting LD_LIBRARY_PATH to .:/mnt/gnu/gcc-3.3/objdir/hppa2.0w-hp-hpux11.11/./lib
gfortran/.libs:/mnt/gnu/gcc-3.3/objdir/gcc:.:/mnt/gnu/gcc-3.3/objdir/hppa2.0w-hp
-hpux11.11/./libgfortran/.libs:/mnt/gnu/gcc-3.3/objdir/gcc
FAIL: gfortran.dg/dev_null.f90  -O0  execution test

Test fails at all optimizations.

-- 
           Summary: FAIL: gfortran.dg/dev_null.f90
           Product: gcc
           Version: 4.0.1
            Status: UNCONFIRMED
          Severity: normal
          Priority: P2
         Component: fortran
        AssignedTo: unassigned at gcc dot gnu dot org
        ReportedBy: danglin at gcc dot gnu dot org
                CC: gcc-bugs at gcc dot gnu dot org
 GCC build triplet: hppa2.0w-hp-hpux11.11
  GCC host triplet: hppa2.0w-hp-hpux11.11
GCC target triplet: hppa2.0w-hp-hpux11.11


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=21593

[Bug fortran/21593] FAIL: gfortran.dg/dev_null.f90

[fxcoudert at gcc dot gnu dot org]

------- Additional Comments From fxcoudert at gcc dot gnu dot org  2005-05-17 08:04 -------
Fails on i386-freebsd, too. Problems with non-linux /dev/null semantic, looks like.

-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|UNCONFIRMED                 |NEW
     Ever Confirmed|                            |1
   Last reconfirmed|0000-00-00 00:00:00         |2005-05-17 08:04:28
               date|                            |


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=21593

[Bug fortran/21593] FAIL: gfortran.dg/dev_null.f90

[fxcoudert at gcc dot gnu dot org]

------- Additional Comments From fxcoudert at gcc dot gnu dot org  2005-05-22 18:20 -------
>From PR21708: "This problem affects FreeBSD, Darwin, and I suspect also
OpenBSD and NetBSD due to their common legacy."

-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=21593

[Bug fortran/21593] FAIL: gfortran.dg/dev_null.f90

[fxcoudert at gcc dot gnu dot org]

------- Additional Comments From fxcoudert at gcc dot gnu dot org  2005-05-22 18:19 -------
*** Bug 21708 has been marked as a duplicate of this bug. ***

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


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=21593

[Bug libfortran/21593] FAIL: gfortran.dg/dev_null.f90

[fxcoudert at gcc dot gnu dot org]

------- Additional Comments From fxcoudert at gcc dot gnu dot org  2005-06-18 08:44 -------
The problem is in the different behavior of ftruncate on /dev/null:

$ cat a.c 
#include <unistd.h>
#include <errno.h>
#include <fcntl.h>
#include <stdio.h>

int main (void)
{
  int fd = open ("/dev/null", O_RDWR);
  ftruncate (fd, 0);
  printf ("%d\n", errno);
  close (fd);
  return 0;
}

On linux, this code outputs 22 (EINVAL), while on freebsd it outputs 0. I
suppose the reason of the failure is similar for other platforms.

I'm sure what to do, but I don't think I'll take too much time thinking about
it, since this a low-priority bug really.

While we wait for this to be solved, I think we should make sure the testcase
only runs on linux and solaris (for which this is known to work).

-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
           Severity|normal                      |minor
   Last reconfirmed|2005-05-17 08:04:28         |2005-06-18 08:44:51
               date|                            |


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=21593

[Bug libfortran/21593] FAIL: gfortran.dg/dev_null.f90

[dave at hiauly1 dot hia dot nrc dot ca]

------- Additional Comments From dave at hiauly1 dot hia dot nrc dot ca  2005-06-18 15:48 -------
Subject: Re:  FAIL: gfortran.dg/dev_null.f90

> On linux, this code outputs 22 (EINVAL), while on freebsd it outputs 0. I
> suppose the reason of the failure is similar for other platforms.

It outputs 0 on hpux 11.11.

Dave


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=21593

[Bug libfortran/21593] FAIL: gfortran.dg/dev_null.f90

[sgk at troutmask dot apl dot washington dot edu]

------- Additional Comments From sgk at troutmask dot apl dot washington dot edu  2005-06-18 16:19 -------
Subject: Re:  FAIL: gfortran.dg/dev_null.f90

> int main (void)
> {
>   int fd = open ("/dev/null", O_RDWR);
>   ftruncate (fd, 0);
>   printf ("%d\n", errno);
>   close (fd);
>   return 0;
> }
> 
> On linux, this code outputs 22 (EINVAL), while on freebsd it outputs 0. I
> suppose the reason of the failure is similar for other platforms.
> 
> I'm sure what to do, but I don't think I'll take too much time thinking about
> it, since this a low-priority bug really.

According to null(4) man page,  reading or writing to /dev/null
on FreeBSD always succeeds.  According to the ftruncate man page,
upon successful completion, the value 0 is returned.

The redhat system that I checked, identifies /dev/null as "a 
special file" in null(4).  The ftruncate man page specifically
states the behavior for "regular files".  So redhat has wiggle
room with its use of errno.

POSIX
http://www.opengroup.org/onlinepubs/009695399/toc.htm
If fildes refers to any other file type, except a shared memory
object, the result is unspecified.

So, we need to check both the return value of ftruncate and
the value of errno.



-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=21593

[Bug libfortran/21593] FAIL: gfortran.dg/dev_null.f90

[dave at hiauly1 dot hia dot nrc dot ca]

------- Additional Comments From dave at hiauly1 dot hia dot nrc dot ca  2005-06-18 16:52 -------
Subject: Re:  FAIL: gfortran.dg/dev_null.f90

> The redhat system that I checked, identifies /dev/null as "a 
> special file" in null(4).  The ftruncate man page specifically
> states the behavior for "regular files".  So redhat has wiggle
> room with its use of errno.

The HP-UX 11.11 manpage says for ftruncate:

      The effect of ftruncate() and truncate() on other types of files is
      unspecified.

The return value from the ftruncate call in the test program is 0.

Dave


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=21593

[Bug libfortran/21593] FAIL: gfortran.dg/dev_null.f90

[kargl at gcc dot gnu dot org]

------- Additional Comments From kargl at gcc dot gnu dot org  2005-07-08 21:01 -------
I've committed FX's mainline change to dev_null.f90 that causes this
test program to only be compiled on linux and solaris to the
4.0 branch.

Andrew, is this sufficient to close this PR?  The behavior of
ftruncate() on /dev/null appears to be system dependent.

-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=21593

[Bug libfortran/21593] FAIL: gfortran.dg/dev_null.f90

[tkoenig at gcc dot gnu dot org]

------- Additional Comments From tkoenig at gcc dot gnu dot org  2005-07-09 14:39 -------
(In reply to comment #8)

> Andrew, is this sufficient to close this PR?  The behavior of
> ftruncate() on /dev/null appears to be system dependent.

Being able to read back from /dev/null is still a bug, no matter
what ftruncate() does.

If ftruncate() does undefined things on special files, we shouldn't
rely on it doing the right thing.

-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=21593

[Bug libfortran/21593] FAIL: gfortran.dg/dev_null.f90

[sgk at troutmask dot apl dot washington dot edu]

------- Additional Comments From sgk at troutmask dot apl dot washington dot edu  2005-07-09 15:28 -------
Subject: Re:  FAIL: gfortran.dg/dev_null.f90

On Sat, Jul 09, 2005 at 02:39:34PM -0000, tkoenig at gcc dot gnu dot org wrote:
> 
> > Andrew, is this sufficient to close this PR?  The behavior of
> > ftruncate() on /dev/null appears to be system dependent.
> 
> Being able to read back from /dev/null is still a bug, no matter
> what ftruncate() does.
> 

It's system dependent in that reading or writing /dev/null
always succeeds (at least on *BSD).  You can't check the
return value or errno to see if there is some failure. 



-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=21593

[Bug libfortran/21593] FAIL: gfortran.dg/dev_null.f90

[tkoenig at gcc dot gnu dot org]

------- Additional Comments From tkoenig at gcc dot gnu dot org  2005-07-10 22:18 -------
The comments in fd_truncate (line 505 ff. in unix.c) read

  if (lseek (s->fd, s->logical_offset, SEEK_SET) == -1)
    return FAILURE;

  /* non-seekable files, like terminals and fifo's fail the lseek.
     the fd is a regular file at this point */


This is clearly wrong for a lot of systems, and needs to be fixed.
This will also fix the /dev/null readback failure.

I'll tackle this.

-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
         AssignedTo|unassigned at gcc dot gnu   |tkoenig at gcc dot gnu dot
                   |dot org                     |org
             Status|NEW                         |ASSIGNED
   Last reconfirmed|2005-06-18 08:44:51         |2005-07-10 22:18:42
               date|                            |


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=21593

[Bug libfortran/21593] FAIL: gfortran.dg/dev_null.f90

[cvs-commit at gcc dot gnu dot org]

------- Additional Comments From cvs-commit at gcc dot gnu dot org  2005-07-12 20:39 -------
Subject: Bug 21593

CVSROOT:	/cvs/gcc
Module name:	gcc
Changes by:	tkoenig@gcc.gnu.org	2005-07-12 20:39:12

Modified files:
	libgfortran    : ChangeLog 
	gcc/testsuite  : ChangeLog 
	gcc/testsuite/gfortran.dg: dev_null.f90 

Log message:
	2005-07-12  Thomas Koenig  <Thomas.Koenig@online.de>
	
	PR libfortran/21593
	gfortran.dg/dev_null.f90:  Remove outdated comment about
	PR libfortran/21593.
	libgfortran/ChangeLog:  Add PR to entry.

Patches:
http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/libgfortran/ChangeLog.diff?cvsroot=gcc&r1=1.262&r2=1.263
http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/testsuite/ChangeLog.diff?cvsroot=gcc&r1=1.5756&r2=1.5757
http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/testsuite/gfortran.dg/dev_null.f90.diff?cvsroot=gcc&r1=1.3&r2=1.4



-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=21593

[Bug libfortran/21593] FAIL: gfortran.dg/dev_null.f90

[tkoenig at gcc dot gnu dot org]

------- Additional Comments From tkoenig at gcc dot gnu dot org  2005-07-12 20:40 -------
Commit for 4.1 here: http://gcc.gnu.org/ml/gcc-cvs/2005-07/msg00486.html

-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=21593

[Bug libfortran/21593] [4.0 only] FAIL: gfortran.dg/dev_null.f90

[pinskia at gcc dot gnu dot org]

-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
            Summary|FAIL:                       |[4.0 only] FAIL:
                   |gfortran.dg/dev_null.f90    |gfortran.dg/dev_null.f90
   Target Milestone|---                         |4.0.2


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=21593

[Bug libfortran/21593] [4.0 only] FAIL: gfortran.dg/dev_null.f90

[cvs-commit at gcc dot gnu dot org]

------- Additional Comments From cvs-commit at gcc dot gnu dot org  2005-07-19 18:03 -------
Subject: Bug 21593

CVSROOT:	/cvs/gcc
Module name:	gcc
Branch: 	gcc-4_0-branch
Changes by:	tkoenig@gcc.gnu.org	2005-07-19 18:03:26

Modified files:
	libgfortran    : ChangeLog 
	libgfortran/io : unix.c 
	gcc/testsuite  : ChangeLog 
	gcc/testsuite/gfortran.dg: dev_null.f90 

Log message:
	2005-07-19  Thomas Koenig  <Thomas.Koenig@online.de>
	
	Backport from mainline.
	PR libfortran/21593
	io/unix.c:  Add member special_file to type unix_stream.
	(fd_truncate):  Don't call ftruncate or chsize if
	s refers to a special file.
	(fd_to_stream):  initialize s->special_file.
	
	2005-07-19  Thomas Koenig  <Thomas.Koenig@online.de>
	
	Backport from mainline.
	PR libfortran/21593
	gfortran.dg/dev_null.f90:  Remove targets.

Patches:
http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/libgfortran/ChangeLog.diff?cvsroot=gcc&only_with_tag=gcc-4_0-branch&r1=1.163.2.66&r2=1.163.2.67
http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/libgfortran/io/unix.c.diff?cvsroot=gcc&only_with_tag=gcc-4_0-branch&r1=1.21.10.7&r2=1.21.10.8
http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/testsuite/ChangeLog.diff?cvsroot=gcc&only_with_tag=gcc-4_0-branch&r1=1.5084.2.288&r2=1.5084.2.289
http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/testsuite/gfortran.dg/dev_null.f90.diff?cvsroot=gcc&only_with_tag=gcc-4_0-branch&r1=1.1.2.2&r2=1.1.2.3



-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=21593

[Bug libfortran/21593] [4.0 only] FAIL: gfortran.dg/dev_null.f90

[tkoenig at gcc dot gnu dot org]

------- Additional Comments From tkoenig at gcc dot gnu dot org  2005-07-19 18:31 -------
Fixed in 4.0.

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


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=21593