[Bug fortran/17535] New: Regression in gfortran with module procedures

[Bug fortran/17535] New: Regression in gfortran with module procedures

[sgk at troutmask dot apl dot washington dot edu]

My m90 math library compiled without a problem on 02 Sep 04.  A 16 Sep 04
gfortran has developed ICEs on at least 4 files.

dhcp-78-96:kargl[206] /home/sgk/bin/gfortran -c sjnm.f90
sjnm.f90:48: internal compiler error: Segmentation fault
Please submit a full bug report,
with preprocessed source if appropriate.
See <URL:http://gcc.gnu.org/bugs.html> for instructions.

-- 
           Summary: Regression in gfortran with module procedures
           Product: gcc
           Version: 4.0.0
            Status: UNCONFIRMED
          Severity: critical
          Priority: P2
         Component: fortran
        AssignedTo: unassigned at gcc dot gnu dot org
        ReportedBy: sgk at troutmask dot apl dot washington dot edu
                CC: gcc-bugs at gcc dot gnu dot org


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=17535

[Bug fortran/17535] Regression in gfortran with module procedures

[sgk at troutmask dot apl dot washington dot edu]

------- Additional Comments From sgk at troutmask dot apl dot washington dot edu  2004-09-17 17:35 -------
Created an attachment (id=7162)
 --> (http://gcc.gnu.org/bugzilla/attachment.cgi?id=7162&action=view)
cut down test case showing the problem.


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=17535

[Bug fortran/17535] Regression in gfortran with module procedures

[sgk at troutmask dot apl dot washington dot edu]

------- Additional Comments From sgk at troutmask dot apl dot washington dot edu  2004-09-17 17:42 -------
The back trace suggets the problem is related to the ChangeLog entry

2004-09-07  Per Bothner  <per@bothner.com>
        Paul Brook  <paul@codesourcery.com>

        * error.c (show_locus): Handle mapped locations.




#0  show_locus (offset=0, loc=0xb2c648) at ../../gcc/gcc/fortran/error.c:129
#1  0x000000000041430f in show_loci (l1=Variable "l1" is not available.
) at ../../gcc/gcc/fortran/error.c:263
#2  0x0000000000413fde in error_print (type=0x851a05 "Error:", 
    format0=0x852a50 "Ambiguous interfaces '%s' and '%s' in %s at %L",
argp=Variable "argp" is not available.
)
    at ../../gcc/gcc/fortran/error.c:374
#3  0x000000000041467e in gfc_error (format=Variable "format" is not available.
)
    at ../../gcc/gcc/fortran/error.c:596
#4  0x0000000000417da7 in check_interface1 (p=0xb2c640, q=0xb2c660, 
    generic_flag=1, interface_name=0x7fffffffdf80 "generic interface 'sj0'")
    at ../../gcc/gcc/fortran/interface.c:935
#5  0x0000000000417e3b in check_sym_interfaces (sym=0xb9d000)
    at ../../gcc/gcc/fortran/interface.c:967
#6  0x000000000044539e in traverse_ns (st=0xb85600, 
    func=0x417dc0 <check_sym_interfaces>)
    at ../../gcc/gcc/fortran/symbol.c:2297
#7  0x0000000000445386 in traverse_ns (st=0xb85300, 
    func=0x417dc0 <check_sym_interfaces>)
    at ../../gcc/gcc/fortran/symbol.c:2300
#8  0x00000000004453dc in gfc_traverse_ns (ns=0xb86000, 
    func=0x417dc0 <check_sym_interfaces>)
    at ../../gcc/gcc/fortran/symbol.c:2314
#9  0x0000000000417f37 in gfc_check_interfaces (ns=0xb86000)
    at ../../gcc/gcc/fortran/interface.c:1016
#10 0x000000000042ea0d in gfc_use_module ()
    at ../../gcc/gcc/fortran/module.c:3129
#11 0x0000000000430c1e in accept_statement (st=Variable "st" is not available.
)
    at ../../gcc/gcc/fortran/parse.c:1015
#12 0x0000000000430e97 in parse_spec (st=ST_USE)
    at ../../gcc/gcc/fortran/parse.c:1583
#13 0x0000000000431eb6 in parse_progunit (st=Variable "st" is not available.
)
    at ../../gcc/gcc/fortran/parse.c:2276
#14 0x0000000000431d9b in parse_contained (module=1)
    at ../../gcc/gcc/fortran/parse.c:2222
#15 0x0000000000432515 in gfc_parse_file ()
    at ../../gcc/gcc/fortran/parse.c:2459
#16 0x00000000004478cb in gfc_be_parse_file (set_yydebug=Variable "set_yydebug"
is not available.



-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=17535

[Bug fortran/17535] Regression in gfortran with module procedures

[pinskia at gcc dot gnu dot org]

-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
           Keywords|                            |ice-on-valid-code


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=17535

[Bug fortran/17535] Regression in gfortran with module procedures

[tobi at gcc dot gnu dot org]

------- Additional Comments From tobi at gcc dot gnu dot org  2004-09-18 13:28 -------
Confirmed. I somehow messed up my CVS tree when trying to back out the patch
Steve is pointing to, so I can't tell for sure if that's the culprit.

-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|UNCONFIRMED                 |NEW
     Ever Confirmed|                            |1
   Last reconfirmed|0000-00-00 00:00:00         |2004-09-18 13:28:09
               date|                            |


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=17535

[Bug fortran/17535] Regression in gfortran with module procedures

[tobi at gcc dot gnu dot org]

------- Additional Comments From tobi at gcc dot gnu dot org  2004-09-18 13:29 -------
Forgot to say: I didn't verify that it actually is a regression, because I don't
have an old build lying around.

-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=17535

[Bug fortran/17535] Regression in gfortran with module procedures

[sgk at troutmask dot apl dot washington dot edu]

------- Additional Comments From sgk at troutmask dot apl dot washington dot edu  2004-09-18 15:00 -------
It's either a latent bug that was exposed by a recent commit or a
regression.  I can assure you that my m90 library compiled without
a problem on Sept 2nd.  In fact, I used gfortran to compile and
run my bio-heat transfer eqaution code (bhte) and my synthetic aperture
sonar proocessing code (tdi) on an amd64 FreeBSD system.  Both programs
use m90 and both programs gave the expected results.  In fact, bhte
was grabbing up to 2.5 GB of memory (we have an inefficient implementation
of WHERE).

-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=17535

[Bug fortran/17535] Regression in gfortran with module procedures

[tobi at gcc dot gnu dot org]

------- Additional Comments From tobi at gcc dot gnu dot org  2004-09-28 17:24 -------
Fixed by reverting: http://gcc.gnu.org/ml/fortran/2004-09/msg00118.html

-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=17535

[Bug fortran/17535] Regression in gfortran with module procedures

[sgk at troutmask dot apl dot washington dot edu]

------- Additional Comments From sgk at troutmask dot apl dot washington dot edu  2004-09-28 20:18 -------
Subject: Re:  Regression in gfortran with module procedures

On Tue, Sep 28, 2004 at 05:24:30PM -0000, tobi at gcc dot gnu dot org wrote:
> 
> Fixed by reverting: http://gcc.gnu.org/ml/fortran/2004-09/msg00118.html
> 

Thanks.



-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=17535

[Bug fortran/17535] [4.0 Regression] gfortran with module procedures

[pinskia at gcc dot gnu dot org]

------- Additional Comments From pinskia at gcc dot gnu dot org  2004-10-07 14:04 -------
Small testcase from PR 17878:
module FOO
  interface BAR

    subroutine BAR1(X)
      integer :: X
    end subroutine

    subroutine BAR2(X)
      real :: X
    end subroutine

  end interface
end module

subroutine BAZ(X)
  use FOO
end subroutine

-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
            Summary|Regression in gfortran with |[4.0 Regression] gfortran
                   |module procedures           |with module procedures
   Target Milestone|---                         |4.0.0


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=17535

[Bug fortran/17535] Regression in gfortran with module procedures

[pinskia at gcc dot gnu dot org]

------- Additional Comments From pinskia at gcc dot gnu dot org  2004-10-07 14:03 -------
*** Bug 17878 has been marked as a duplicate of this bug. ***

-- 
           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |sfilippone at uniroma2 dot
                   |                            |it


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=17535

[Bug fortran/17535] [4.0 Regression] gfortran with module procedures

[tobi at gcc dot gnu dot org]

------- Additional Comments From tobi at gcc dot gnu dot org  2004-10-09 16:23 -------
With the latest fix to module.c the reduced testcase now gives these erroneous
errors:

[tobi@marktplatz tests]$ ~/src/gcc/build-clean/gcc/f951 pr17535.f90
 In file pr17535.f90:16

  use FOO
        1
Error: Ambiguous interfaces 'bar2' and 'bar1' in generic interface 'bar' at (1)
 In file pr17535.f90:16

  use FOO
        1
Error: Ambiguous interfaces 'bar2' and 'bar1' in generic interface 'bar' at (1)

Execution times (seconds)
 parser                :   0.01 (50%) usr   0.01 (50%) sys   0.24 (19%) wall
 symout                :   0.00 ( 0%) usr   0.00 ( 0%) sys   0.02 ( 2%) wall
 TOTAL                 :   0.02             0.02             1.28
Extra diagnostic checks enabled; compiler may run slowly.
Configure with --disable-checking to disable checks.


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=17535

[Bug fortran/17535] [4.0 Regression] gfortran with module procedures

[pinskia at gcc dot gnu dot org]

------- Additional Comments From pinskia at gcc dot gnu dot org  2004-10-10 18:14 -------
Patch here: <http://gcc.gnu.org/ml/gcc-patches/2004-10/msg00856.html>.

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


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=17535

[Bug fortran/17535] [4.0 Regression] gfortran with module procedures

[cvs-commit at gcc dot gnu dot org]

------- Additional Comments From cvs-commit at gcc dot gnu dot org  2004-11-03 00:54 -------
Subject: Bug 17535

CVSROOT:	/cvs/gcc
Module name:	gcc
Changes by:	tobi@gcc.gnu.org	2004-11-03 00:54:02

Modified files:
	gcc/fortran    : ChangeLog module.c 
	gcc/testsuite  : ChangeLog 
Added files:
	gcc/testsuite/gfortran.dg: generic_1.f90 generic_2.f90 
	                           generic_3.f90 

Log message:
	fortran/
	PR fortran/17535
	PR fortran/17583
	PR fortran/17713
	* module.c (write_symbol1): Set module_name for dummy arguments.
	
	testsuite/
	PR fortran/17535
	PR fortran/17583
	PR fortran/17713
	* gfortran.dg/generic_[123].f90: New testcases.

Patches:
http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/fortran/ChangeLog.diff?cvsroot=gcc&r1=1.246&r2=1.247
http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/fortran/module.c.diff?cvsroot=gcc&r1=1.19&r2=1.20
http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/testsuite/ChangeLog.diff?cvsroot=gcc&r1=1.4536&r2=1.4537
http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/testsuite/gfortran.dg/generic_1.f90.diff?cvsroot=gcc&r1=NONE&r2=1.1
http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/testsuite/gfortran.dg/generic_2.f90.diff?cvsroot=gcc&r1=NONE&r2=1.1
http://gcc.gnu.org/cgi-bin/cvsweb.cgi/gcc/gcc/testsuite/gfortran.dg/generic_3.f90.diff?cvsroot=gcc&r1=NONE&r2=1.1



-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=17535

[Bug fortran/17535] [4.0 Regression] gfortran with module procedures

[tobi at gcc dot gnu dot org]

------- Additional Comments From tobi at gcc dot gnu dot org  2004-11-03 00:57 -------
Fixed.

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


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=17535