* [Bug fortran/54350] [4.8 Regression] FAIL: gfortran.dg/realloc_on_assign_*.f90 -O (internal compiler error) at r190586
2012-08-22 16:25 [Bug fortran/54350] New: [4.8 Regression] FAIL: gfortran.dg/realloc_on_assign_*.f90 -O (internal compiler error) at r190586 dominiq at lps dot ens.fr
@ 2012-08-22 20:00 ` hp at gcc dot gnu.org
2012-08-22 20:50 ` burnus at gcc dot gnu.org
` (3 subsequent siblings)
4 siblings, 0 replies; 6+ messages in thread
From: hp at gcc dot gnu.org @ 2012-08-22 20:00 UTC (permalink / raw)
To: gcc-bugs
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=54350
Hans-Peter Nilsson <hp at gcc dot gnu.org> changed:
What |Removed |Added
----------------------------------------------------------------------------
Status|UNCONFIRMED |NEW
Last reconfirmed| |2012-08-22
CC| |burnus at gcc dot gnu.org,
| |hp at gcc dot gnu.org
Ever Confirmed|0 |1
--- Comment #1 from Hans-Peter Nilsson <hp at gcc dot gnu.org> 2012-08-22 19:59:16 UTC ---
Seeing something like this for cris-elf too, specifically a regression for
gfortran.sum gfortran.dg/realloc_on_assign_14.f90
gfortran.sum gfortran.dg/realloc_on_assign_15.f90
gfortran.sum gfortran.dg/realloc_on_assign_2.f03
for exactly r190586 except in my case (cross from x86_64-linux to cris-elf) the
ICE message is the less specific:
/tmp/hpautotest-gcc0/gcc/gcc/testsuite/gfortran.dg/realloc_on_assign_14.f90: In
function 'MAIN__':
/tmp/hpautotest-gcc0/gcc/gcc/testsuite/gfortran.dg/realloc_on_assign_14.f90:31:0:
internal compiler error: Segmentation fault
Adding committers' @gcc.gnu.org address to CC.
^ permalink raw reply [flat|nested] 6+ messages in thread* [Bug fortran/54350] [4.8 Regression] FAIL: gfortran.dg/realloc_on_assign_*.f90 -O (internal compiler error) at r190586
2012-08-22 16:25 [Bug fortran/54350] New: [4.8 Regression] FAIL: gfortran.dg/realloc_on_assign_*.f90 -O (internal compiler error) at r190586 dominiq at lps dot ens.fr
2012-08-22 20:00 ` [Bug fortran/54350] " hp at gcc dot gnu.org
@ 2012-08-22 20:50 ` burnus at gcc dot gnu.org
2012-08-24 7:44 ` burnus at gcc dot gnu.org
` (2 subsequent siblings)
4 siblings, 0 replies; 6+ messages in thread
From: burnus at gcc dot gnu.org @ 2012-08-22 20:50 UTC (permalink / raw)
To: gcc-bugs
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=54350
Tobias Burnus <burnus at gcc dot gnu.org> changed:
What |Removed |Added
----------------------------------------------------------------------------
Status|NEW |ASSIGNED
AssignedTo|unassigned at gcc dot |burnus at gcc dot gnu.org
|gnu.org |
--- Comment #2 from Tobias Burnus <burnus at gcc dot gnu.org> 2012-08-22 20:49:08 UTC ---
Yes, that's a side-effect of my recent patch - sorry. I know about the issue
since this morning, but I didn't manage to fix it as quickly as I had hoped
for. Debugging memory leaks and understanding the scalarizer took longer than
anticipated.
The simplest is to revert
http://gcc.gnu.org/ml/fortran/2012-08/msg00120.html
Some issues and a possible approach to fix them is given in
http://gcc.gnu.org/ml/fortran/2012-08/msg00131.html
and the reply
http://gcc.gnu.org/ml/fortran/2012-08/msg00133.html
I have a patch which is currently building, but with build fixes, other
changes, and regtesting it will probably take a while until it is ready.
^ permalink raw reply [flat|nested] 6+ messages in thread* [Bug fortran/54350] [4.8 Regression] FAIL: gfortran.dg/realloc_on_assign_*.f90 -O (internal compiler error) at r190586
2012-08-22 16:25 [Bug fortran/54350] New: [4.8 Regression] FAIL: gfortran.dg/realloc_on_assign_*.f90 -O (internal compiler error) at r190586 dominiq at lps dot ens.fr
2012-08-22 20:00 ` [Bug fortran/54350] " hp at gcc dot gnu.org
2012-08-22 20:50 ` burnus at gcc dot gnu.org
@ 2012-08-24 7:44 ` burnus at gcc dot gnu.org
2012-08-24 7:46 ` burnus at gcc dot gnu.org
2012-08-24 7:46 ` burnus at gcc dot gnu.org
4 siblings, 0 replies; 6+ messages in thread
From: burnus at gcc dot gnu.org @ 2012-08-24 7:44 UTC (permalink / raw)
To: gcc-bugs
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=54350
--- Comment #3 from Tobias Burnus <burnus at gcc dot gnu.org> 2012-08-24 07:43:31 UTC ---
Author: burnus
Date: Fri Aug 24 07:43:23 2012
New Revision: 190641
URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=190641
Log:
2012-08-24 Tobias Burnus <burnus@net-b.de>
PR fortran/54350
* trans-array.c (free_ss_info): Free data.array.subscript.
(gfc_free_ss): No longer free data.array.subscript.
(walk_coarray): New function, moved from trans-intrinsic.c
(gfc_conv_expr_descriptor): Walk array descriptor instead
of taking passed "ss".
(get_array_ctor_all_strlen, gfc_add_loop_ss_code,
gfc_conv_array_parameter): Update call and cleanup ss handling.
* trans-array.h (gfc_conv_expr_descriptor,
gfc_conv_array_parameter): Update prototype.
* trans-expr.c (gfc_conv_derived_to_class,
conv_isocbinding_procedure, gfc_conv_procedure_call,
gfc_trans_alloc_subarray_assign, gfc_trans_subcomponent_assign,
gfc_trans_pointer_assignment, gfc_trans_arrayfunc_assign): Update
call to gfc_conv_expr_descriptor and gfc_conv_array_parameter,
and clean up.
* trans-intrinsic.c (walk_coarray): Moved to trans-array.c
(trans_this_image, trans_image_index, gfc_conv_intrinsic_rank
gfc_conv_intrinsic_bound, gfc_conv_intrinsic_cobound,
gfc_conv_intrinsic_len, gfc_conv_intrinsic_size,
gfc_conv_intrinsic_sizeof, gfc_conv_intrinsic_storage_size,
gfc_conv_intrinsic_transfer, gfc_conv_allocated,
gfc_conv_associated, gfc_conv_intrinsic_loc,
conv_intrinsic_move_alloc): Update calls.
* trans-io.c (gfc_convert_array_to_string, set_internal_unit,
gfc_trans_transfer): Ditto.
* trans-stmt.c (gfc_conv_elemental_dependencies,
gfc_trans_sync, trans_associate_var,
gfc_trans_pointer_assign_need_temp): Ditto.
Modified:
trunk/gcc/fortran/ChangeLog
trunk/gcc/fortran/trans-array.c
trunk/gcc/fortran/trans-array.h
trunk/gcc/fortran/trans-expr.c
trunk/gcc/fortran/trans-intrinsic.c
trunk/gcc/fortran/trans-io.c
trunk/gcc/fortran/trans-stmt.c
^ permalink raw reply [flat|nested] 6+ messages in thread* [Bug fortran/54350] [4.8 Regression] FAIL: gfortran.dg/realloc_on_assign_*.f90 -O (internal compiler error) at r190586
2012-08-22 16:25 [Bug fortran/54350] New: [4.8 Regression] FAIL: gfortran.dg/realloc_on_assign_*.f90 -O (internal compiler error) at r190586 dominiq at lps dot ens.fr
` (2 preceding siblings ...)
2012-08-24 7:44 ` burnus at gcc dot gnu.org
@ 2012-08-24 7:46 ` burnus at gcc dot gnu.org
2012-08-24 7:46 ` burnus at gcc dot gnu.org
4 siblings, 0 replies; 6+ messages in thread
From: burnus at gcc dot gnu.org @ 2012-08-24 7:46 UTC (permalink / raw)
To: gcc-bugs
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=54350
Tobias Burnus <burnus at gcc dot gnu.org> changed:
What |Removed |Added
----------------------------------------------------------------------------
Status|ASSIGNED |RESOLVED
Resolution| |FIXED
--- Comment #3 from Tobias Burnus <burnus at gcc dot gnu.org> 2012-08-24 07:43:31 UTC ---
Author: burnus
Date: Fri Aug 24 07:43:23 2012
New Revision: 190641
URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=190641
Log:
2012-08-24 Tobias Burnus <burnus@net-b.de>
PR fortran/54350
* trans-array.c (free_ss_info): Free data.array.subscript.
(gfc_free_ss): No longer free data.array.subscript.
(walk_coarray): New function, moved from trans-intrinsic.c
(gfc_conv_expr_descriptor): Walk array descriptor instead
of taking passed "ss".
(get_array_ctor_all_strlen, gfc_add_loop_ss_code,
gfc_conv_array_parameter): Update call and cleanup ss handling.
* trans-array.h (gfc_conv_expr_descriptor,
gfc_conv_array_parameter): Update prototype.
* trans-expr.c (gfc_conv_derived_to_class,
conv_isocbinding_procedure, gfc_conv_procedure_call,
gfc_trans_alloc_subarray_assign, gfc_trans_subcomponent_assign,
gfc_trans_pointer_assignment, gfc_trans_arrayfunc_assign): Update
call to gfc_conv_expr_descriptor and gfc_conv_array_parameter,
and clean up.
* trans-intrinsic.c (walk_coarray): Moved to trans-array.c
(trans_this_image, trans_image_index, gfc_conv_intrinsic_rank
gfc_conv_intrinsic_bound, gfc_conv_intrinsic_cobound,
gfc_conv_intrinsic_len, gfc_conv_intrinsic_size,
gfc_conv_intrinsic_sizeof, gfc_conv_intrinsic_storage_size,
gfc_conv_intrinsic_transfer, gfc_conv_allocated,
gfc_conv_associated, gfc_conv_intrinsic_loc,
conv_intrinsic_move_alloc): Update calls.
* trans-io.c (gfc_convert_array_to_string, set_internal_unit,
gfc_trans_transfer): Ditto.
* trans-stmt.c (gfc_conv_elemental_dependencies,
gfc_trans_sync, trans_associate_var,
gfc_trans_pointer_assign_need_temp): Ditto.
Modified:
trunk/gcc/fortran/ChangeLog
trunk/gcc/fortran/trans-array.c
trunk/gcc/fortran/trans-array.h
trunk/gcc/fortran/trans-expr.c
trunk/gcc/fortran/trans-intrinsic.c
trunk/gcc/fortran/trans-io.c
trunk/gcc/fortran/trans-stmt.c
--- Comment #4 from Tobias Burnus <burnus at gcc dot gnu.org> 2012-08-24 07:44:55 UTC ---
FIXED.
Sorry for the breakage, which took a bit longer than expected to fix. At least,
gfortran now leaks much less gfc_ss structs then it did before.
^ permalink raw reply [flat|nested] 6+ messages in thread* [Bug fortran/54350] [4.8 Regression] FAIL: gfortran.dg/realloc_on_assign_*.f90 -O (internal compiler error) at r190586
2012-08-22 16:25 [Bug fortran/54350] New: [4.8 Regression] FAIL: gfortran.dg/realloc_on_assign_*.f90 -O (internal compiler error) at r190586 dominiq at lps dot ens.fr
` (3 preceding siblings ...)
2012-08-24 7:46 ` burnus at gcc dot gnu.org
@ 2012-08-24 7:46 ` burnus at gcc dot gnu.org
4 siblings, 0 replies; 6+ messages in thread
From: burnus at gcc dot gnu.org @ 2012-08-24 7:46 UTC (permalink / raw)
To: gcc-bugs
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=54350
Tobias Burnus <burnus at gcc dot gnu.org> changed:
What |Removed |Added
----------------------------------------------------------------------------
Status|ASSIGNED |RESOLVED
Resolution| |FIXED
--- Comment #4 from Tobias Burnus <burnus at gcc dot gnu.org> 2012-08-24 07:44:55 UTC ---
FIXED.
Sorry for the breakage, which took a bit longer than expected to fix. At least,
gfortran now leaks much less gfc_ss structs then it did before.
^ permalink raw reply [flat|nested] 6+ messages in thread