Re: [PATCH] fortran PR95542 Fix ICE for deferred-length CHARACTER, result via host association

On 8/16/26 12:02 PM, Mikael Morin wrote:
> Le 16/08/2026 à 02:29, Jerry D a écrit :
>> Attached is the patch I re-discovered while cleaning up worktrees.
>>
>> Regression tested on x86_64.
>>
>> OK for mainline?
>>
>> Regards,
>>
>> Jerry
>> ---
> 
>>
>> gfc_get_symbol_decl has an assert meant to verify that a deferred-length
>> string's length variable lives in the same scope as the symbol it
>> belongs to.  This is wrong when sym->backend_decl is itself the enclosing
>> function's FUNCTION_DECL.  The correct comparison is
>> DECL_CONTEXT (length) == sym->backend_decl directly.
>>
> 
>> diff --git a/gcc/fortran/trans-decl.cc b/gcc/fortran/trans-decl.cc
>> index 47b28c1d003..df5aef2fd86 100644
>> --- a/gcc/fortran/trans-decl.cc
>> +++ b/gcc/fortran/trans-decl.cc
>> @@ -1848,10 +1848,15 @@ gfc_get_symbol_decl (gfc_symbol * sym)
>>              gfc_add_decl_to_parent_function (length);
>>          }
>>
>> +          /* When the symbol's own backend_decl is a FUNCTION_DECL, its
>> +         DECL_CONTEXT is where that function itself is declared, not
>> +         where its locals live.  */
>>            gcc_assert (sym->backend_decl == current_function_decl
>>                ? DECL_CONTEXT (length) == current_function_decl
>> -              : (DECL_CONTEXT (sym->backend_decl)
>> -                 == DECL_CONTEXT (length)));
>> +              : (TREE_CODE (sym->backend_decl) == FUNCTION_DECL
>> +                 ? DECL_CONTEXT (length) == sym->backend_decl
>> +                 : (DECL_CONTEXT (sym->backend_decl)
>> +                == DECL_CONTEXT (length))));
> The first condition (sym->backend_decl == current_function_decl) is redundant 
> with the second one (TREE_CODE (sym->backend_decl) == FUNCTION_DECL); I think 
> the outer conditional can be removed completely.
> 
> OK with that change.

Agree, thanks for review, edit made. I will commit after a complete testsuite check

Thanks again,

Jerry
---