[Bug fortran/84868] [9/10/11/12 Regression] ICE in gfc_conv_descriptor_offset, at fortran/trans-array.c:208

[Bug fortran/84868] [9/10/11/12 Regression] ICE in gfc_conv_descriptor_offset, at fortran/trans-array.c:208

[jakub at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=84868

Jakub Jelinek <jakub at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
   Target Milestone|8.5                         |9.4

--- Comment #9 from Jakub Jelinek <jakub at gcc dot gnu.org> ---
GCC 8 branch is being closed.

[Bug fortran/84868] [9/10/11/12 Regression] ICE in gfc_conv_descriptor_offset, at fortran/trans-array.c:208

[rguenth at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=84868

Richard Biener <rguenth at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
   Target Milestone|9.4                         |9.5

--- Comment #10 from Richard Biener <rguenth at gcc dot gnu.org> ---
GCC 9.4 is being released, retargeting bugs to GCC 9.5.

[Bug fortran/84868] [10/11/12/13 Regression] ICE in gfc_conv_descriptor_offset, at fortran/trans-array.c:208

[rguenth at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=84868

Richard Biener <rguenth at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
   Target Milestone|9.5                         |10.4

--- Comment #11 from Richard Biener <rguenth at gcc dot gnu.org> ---
GCC 9 branch is being closed

[Bug fortran/84868] [10/11/12/13 Regression] ICE in gfc_conv_descriptor_offset, at fortran/trans-array.c:208

[jakub at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=84868

Jakub Jelinek <jakub at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
   Target Milestone|10.4                        |10.5

--- Comment #12 from Jakub Jelinek <jakub at gcc dot gnu.org> ---
GCC 10.4 is being released, retargeting bugs to GCC 10.5.

[Bug fortran/84868] [11/12/13/14 Regression] ICE in gfc_conv_descriptor_offset, at fortran/trans-array.c:208

[rguenth at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=84868

Richard Biener <rguenth at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
   Target Milestone|10.5                        |11.5

--- Comment #13 from Richard Biener <rguenth at gcc dot gnu.org> ---
GCC 10 branch is being closed.

[Bug fortran/84868] [11/12/13/14 Regression] ICE in gfc_conv_descriptor_offset, at fortran/trans-array.c:208

[pault at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=84868

Paul Thomas <pault at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
           Assignee|unassigned at gcc dot gnu.org      |pault at gcc dot gnu.org
                 CC|                            |pault at gcc dot gnu.org

--- Comment #14 from Paul Thomas <pault at gcc dot gnu.org> ---
(In reply to G. Steinmetz from comment #0)
> Changed between 20161204 and 20161211 :
> 
> 
> $ cat z1.f90
> module m
>    character(:), allocatable :: c
> contains
>    function f(n) result(z)
>       character, parameter :: c(3) = ['x', 'y', 'z']
>       integer, intent(in) :: n
>       character(len_trim(c(n))) :: z
>       z = c(n)
>    end
> end
> program p
>    use m
>    print *, f(2)
> end
> 
> 
> $ gfortran-7-20161204 -c z1.f90
> $
> $ gfortran-8-20180311 -c z1.f90
> z1.f90:13:0:
> 
>     print *, f(2)
> 
> internal compiler error: in gfc_conv_descriptor_offset, at
> fortran/trans-array.c:208
> 0x74e222 gfc_conv_descriptor_offset
>         ../../gcc/fortran/trans-array.c:208
> 0x75349c gfc_conv_descriptor_offset_get(tree_node*)
>         ../../gcc/fortran/trans-array.c:220
> 0x75349c gfc_conv_array_offset(tree_node*)
>         ../../gcc/fortran/trans-array.c:2967
> 0x75349c gfc_conv_array_ref(gfc_se*, gfc_array_ref*, gfc_expr*, locus*)
>         ../../gcc/fortran/trans-array.c:3575
> 0x780ddd gfc_conv_variable
>         ../../gcc/fortran/trans-expr.c:2737
> 0x77db02 gfc_conv_expr(gfc_se*, gfc_expr*)
>         ../../gcc/fortran/trans-expr.c:7926
> 0x78a98c gfc_conv_intrinsic_function_args
>         ../../gcc/fortran/trans-intrinsic.c:223
> 0x79dc3d gfc_conv_intrinsic_len_trim
>         ../../gcc/fortran/trans-intrinsic.c:6248
> 0x79dc3d gfc_conv_intrinsic_function(gfc_se*, gfc_expr*)
>         ../../gcc/fortran/trans-intrinsic.c:9134
> 0x77d545 gfc_conv_function_expr
>         ../../gcc/fortran/trans-expr.c:6784
> 0x77dae2 gfc_conv_expr(gfc_se*, gfc_expr*)
>         ../../gcc/fortran/trans-expr.c:7918
> 0x77fa8a gfc_apply_interface_mapping(gfc_interface_mapping*, gfc_se*,
> gfc_expr*)
>         ../../gcc/fortran/trans-expr.c:4409
> 0x77b6f7 gfc_conv_procedure_call(gfc_se*, gfc_symbol*, gfc_actual_arglist*,
> gfc_expr*, vec<tree_node*, va_gc, vl_embed>*)
>         ../../gcc/fortran/trans-expr.c:5970
> 0x77d59c gfc_conv_function_expr
>         ../../gcc/fortran/trans-expr.c:6808
> 0x77dae2 gfc_conv_expr(gfc_se*, gfc_expr*)
>         ../../gcc/fortran/trans-expr.c:7918
> 0x7843aa gfc_conv_expr_reference(gfc_se*, gfc_expr*)
>         ../../gcc/fortran/trans-expr.c:8018
> 0x7a3d56 gfc_trans_transfer(gfc_code*)
>         ../../gcc/fortran/trans-io.c:2585
> 0x749ec7 trans_code
>         ../../gcc/fortran/trans.c:2044
> 0x7a1807 build_dt
>         ../../gcc/fortran/trans-io.c:2027
> 0x749ee7 trans_code
>         ../../gcc/fortran/trans.c:2016

This, I believe, was commit 345bd7ebbb38f0e1d5acf33ab3f680111cfa7871 where
LEN_TRIM was introduced to interface mapping on 2016-12-09 for pr44265.
Removing the chunk concerned does not fix the problem.

I have stared at this for ages. For reasons that I cannot see, the use
associated version of 'c' is being used, so the interface mapping has failed.

I will come back to this.

Paul

[Bug fortran/84868] [11/12/13/14/15 Regression] ICE in gfc_conv_descriptor_offset, at fortran/trans-array.c:208

[pault at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=84868

--- Comment #15 from Paul Thomas <pault at gcc dot gnu.org> ---
(In reply to Harald Anlauf from comment #6)
> (In reply to Harald Anlauf from comment #5)
> > With the following patch len_trim is accepted in a specification expression:
> 
> Just forget that.

Indeed:
NAG Fortran Compiler Release 7.1(Hanzomon) Build 7133
Error: orig.f90, line 7: C is not permitted in a specification expression

[pault@pc30 pr84868]$ ifort o*.f90 ;./a.out
orig.f90(5): error #6219: This variable, used in a specification expression,
must be a dummy argument, a COMMON block object, or an object accessible
through host or use association.   [C]

Paul

[Bug fortran/84868] [11/12/13/14/15 Regression] ICE in gfc_conv_descriptor_offset, at fortran/trans-array.c:208

[pault at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=84868

--- Comment #16 from Paul Thomas <pault at gcc dot gnu.org> ---
Created attachment 58641
  --> https://gcc.gnu.org/bugzilla/attachment.cgi?id=58641&action=edit
Fix for this PR

The only way that I have found to fix this is to simplify the len_trim
expression. In doing so, another parameter has to be created for which I have
taken a leaf out of the unique_name mechanism. Unfortunately, two such
parameters are created (parsing specification expression and then resolution),
the first one of which is unused. I don't see any easy way to eliminate it.

I came across another bug in testing variants, which is also fixed. Regtests
OK.

I'll think further about doubling up on the parameter before submitting.

Paul

[Bug fortran/84868] [11/12/13/14/15 Regression] ICE in gfc_conv_descriptor_offset, at fortran/trans-array.c:208

[pault at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=84868

--- Comment #17 from Paul Thomas <pault at gcc dot gnu.org> ---
(In reply to Paul Thomas from comment #16)
> Created attachment 58641 [details]
> Fix for this PR
> 
> The only way that I have found to fix this is to simplify the len_trim
> expression. In doing so, another parameter has to be created for which I
> have taken a leaf out of the unique_name mechanism. Unfortunately, two such
> parameters are created (parsing specification expression and then
> resolution), the first one of which is unused. I don't see any easy way to
> eliminate it.
> 
> I came across another bug in testing variants, which is also fixed. Regtests
> OK.
> 
> I'll think further about doubling up on the parameter before submitting.
> 
> Paul

I have figured out how to eliminate the duplicate parameter. I will be
submitting to the list very soon :-)

Cheers

Paul

[Bug fortran/84868] [11/12/13/14/15 Regression] ICE in gfc_conv_descriptor_offset, at fortran/trans-array.c:208

[cvs-commit at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=84868

--- Comment #18 from GCC Commits <cvs-commit at gcc dot gnu.org> ---
The master branch has been updated by Paul Thomas <pault@gcc.gnu.org>:

https://gcc.gnu.org/g:9f966b6a8ff0244dd6f8bf36d876799d5f9bbaee

commit r15-2072-g9f966b6a8ff0244dd6f8bf36d876799d5f9bbaee
Author: Paul Thomas <pault@gcc.gnu.org>
Date:   Tue Jul 16 15:56:44 2024 +0100

    Fortran: Simplify len_trim with array ref and fix mapping bug[PR84868].

    2024-07-16  Paul Thomas  <pault@gcc.gnu.org>

    gcc/fortran
            PR fortran/84868
            * simplify.cc (gfc_simplify_len_trim): If the argument is an
            element of a parameter array, simplify all the elements and
            build a new parameter array to hold the result, after checking
            that it doesn't already exist.
            * trans-expr.cc (gfc_get_interface_mapping_array) if a string
            length is available, use it for the typespec.
            (gfc_add_interface_mapping): Supply the se string length.

    gcc/testsuite/
            PR fortran/84868
            * gfortran.dg/pr84868.f90: New test.

[Bug fortran/84868] [12/13/14 Regression] ICE in gfc_conv_descriptor_offset, at fortran/trans-array.c:208

[rguenth at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=84868

Richard Biener <rguenth at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
   Target Milestone|11.5                        |12.5

--- Comment #19 from Richard Biener <rguenth at gcc dot gnu.org> ---
GCC 11 branch is being closed.