[Bug fortran/25576] New: [4.0 only] checking failure in execute/intrinsic_eoshift.f90

[Bug fortran/25576] New: [4.0 only] checking failure in execute/intrinsic_eoshift.f90

[ghazi at gcc dot gnu dot org]

I'm getting a gfortran failure which only appears with checking turned on the
gcc-4.0.x branch:
http://gcc.gnu.org/ml/gcc-testresults/2005-12/msg00417.html

It does not happen in 4.1/mainline.  It does not happen if I target i686.

To repeat configure with --target=x86_64-unknown-linux-gnu
--enable-checking=yes,rtl


-- 
           Summary: [4.0 only] checking failure in
                    execute/intrinsic_eoshift.f90
           Product: gcc
           Version: 4.0.3
            Status: UNCONFIRMED
          Keywords: ice-checking
          Severity: normal
          Priority: P3
         Component: fortran
        AssignedTo: unassigned at gcc dot gnu dot org
        ReportedBy: ghazi at gcc dot gnu dot org
GCC target triplet: x86_64-unknown-linux-gnu


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=25576

[Bug fortran/25576] [4.0 only] checking failure in execute/intrinsic_eoshift.f90

[ghazi at gcc dot gnu dot org]

------- Comment #1 from ghazi at gcc dot gnu dot org  2005-12-27 15:22 -------
Running f951 under gdb, I get:

(gdb) run 
/tmp/kg/40/egcc-4.0-SVN20051226/gcc/testsuite/gfortran.fortran-torture/execute/intrinsic_eoshift.f90
-quiet -dumpbase intrinsic_eoshift.f90 -mtune=k8 -auxbase intrinsic_eoshift -O3
-w -version -fomit-frame-pointer -funroll-loops -o intrinsic_eoshift.s

Program received signal SIGSEGV, Segmentation fault.
reorder_insns (from=0x2aaaae0ad940, to=0x2aaaae0ad940, after=0x2aaaae0ad980)
    at ../../egcc-4.0-SVN20051226/gcc/emit-rtl.c:3649
3649          bb->flags |= BB_DIRTY;

(gdb) p *bb
Cannot access memory at address 0x2aaaad383820


-- 

ghazi at gcc dot gnu dot org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
            Summary|[4.0 only] checking failure |[4.0 only] checking failure
                   |in                          |in
                   |execute/intrinsic_eoshift.f9|execute/intrinsic_eoshift.f9
                   |0                           |0


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=25576

[Bug fortran/25576] [4.0 only] checking failure in execute/intrinsic_eoshift.f90

[ghazi at gcc dot gnu dot org]

------- Comment #2 from ghazi at gcc dot gnu dot org  2005-12-27 15:37 -------
The particular checking type exposing the failure is "gc" checking.  That
forces the garbage collection parameters to be:
--param ggc-min-expand=30 --param ggc-min-heapsize=4096

Using those parameters to compile the testcase in a totally non-checking
bootstrap is enough to trigger the failure.

FYI, the default GC parameters on my box without checking are:
--param ggc-min-expand=100 --param ggc-min-heapsize=131072

I'm guessing that the lower parameters cause us to run a collection which
stomping on something used here.


-- 

ghazi at gcc dot gnu dot org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
           Keywords|                            |GC


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=25576

[Bug fortran/25576] [4.0 only] checking failure in execute/intrinsic_eoshift.f90

[pinskia at gcc dot gnu dot org]

-- 

pinskia at gcc dot gnu dot org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
           Severity|normal                      |minor
   Target Milestone|---                         |4.0.3


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=25576

[Bug fortran/25576] [4.0 only] checking failure in execute/intrinsic_eoshift.f90

[mmitchel at gcc dot gnu dot org]

------- Comment #3 from mmitchel at gcc dot gnu dot org  2006-02-27 20:33 -------
Fortran is not release-critical; P5.


-- 

mmitchel at gcc dot gnu dot org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
           Priority|P3                          |P5


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=25576

[Bug fortran/25576] [4.0 only] checking failure in execute/intrinsic_eoshift.f90

[mmitchel at gcc dot gnu dot org]

-- 

mmitchel at gcc dot gnu dot org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
   Target Milestone|4.0.3                       |4.0.4


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=25576

[Bug fortran/25576] [4.0 only] checking failure in execute/intrinsic_eoshift.f90

[fxcoudert at gcc dot gnu dot org]

------- Comment #4 from fxcoudert at gcc dot gnu dot org  2006-06-07 14:35 -------
This is only on 4.0, and with non-default options, so I'm closing this as
WONTFIX. gfortran is helplessly broken on 4.0 anyway.


-- 

fxcoudert at gcc dot gnu dot org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |fxcoudert at gcc dot gnu dot
                   |                            |org
             Status|UNCONFIRMED                 |RESOLVED
         Resolution|                            |WONTFIX


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=25576