[Patch, fortran] PR fortran/92142 - CFI_setpointer corrupts descriptor

[Mark Eggleston]

Unfortunately ISO_Fortran_binding_16.f90 contains a typo resulting in:

FAIL: gfortran.dg/ISO_Fortran_binding_16.f90Â Â  -O0Â  (test for excess errors)
FAIL: gfortran.dg/ISO_Fortran_binding_16.f90Â Â  -O1Â  (test for excess errors)
FAIL: gfortran.dg/ISO_Fortran_binding_16.f90Â Â  -O2Â  (test for excess errors)
FAIL: gfortran.dg/ISO_Fortran_binding_16.f90Â Â  -O3 -fomit-frame-pointer 
-funroll-loops -fpeel-loops -ftracer -finline-functions  (test for 
excess errors)
FAIL: gfortran.dg/ISO_Fortran_binding_16.f90Â Â  -O3 -g (test for excess 
errors)
FAIL: gfortran.dg/ISO_Fortran_binding_16.f90   -Os  (test for excess errors)

the cause is that it refers to ISO_Fortran_binding_15.c instead of 
ISO_Fortran_binding_16.c in the dg-additional-sources directive.

regards,

Mark

On 11/11/2019 10:29, Tobias Burnus wrote:
> Hi José,
>
> On 10/29/19 11:35 AM, José Rui Faustino de Sousa wrote:
>> Added the suggested check and a few touches to comments and the error 
>> message.
> Thanks.
>>> that might be a bug in the C code of the test itself
>> I took a look and although there are problems with the code of the 
>> test I do not think they are relevant.
>>
>> Index: gcc/testsuite/gfortran.dg/ISO_Fortran_binding_10.c
>
> I have included that patch.
>
> Other remark: ChangeLog missing in the updated version of the patch. 
> (I just saw that you did include it when posting initially.)
>
> I have now committed the patch as Rev. 278048 to the trunk.
> Thanks for the patch!
>
> Cheers,
>
> Tobias
>
-- 
https://www.codethink.co.uk/privacy.html