[gfortran,commited] Fix a bunch of dg directives in the testsuite

FX Coudert fxcoudert@gmail.com
Thu Mar 29 21:57:00 GMT 2007

Hi all,

I had for some time a patch to fix some directives in the testsuite,  
as some of them were bogus ("dg-run" instead of "dg-do run", for  
example; in that particular case, the testcase was actually never  
run). Commited after regtesting on i686-linux. (see below)

One case is troubling, though: gfortran.dg/pointer_intent_1.f90 has a  
"dg-run" instead of "dg-do run", and when I fixed that, it is run and  
I discovered it fails at -O0. Tobias, could you look into it? Or do  
you prefer that I open a PR?

(PS: for some reason, line 63 of gfortran.dg/pointer_intent_1.f90 is  
commented out: "!if(comp%point /=  27) call abort()". Would you mind  
adding a comment about why it is so? I failed to guess the reason...)



2007-03-29  Francois-Xavier Coudert  <fxcoudert@gcc.gnu.org>

         * size_dim.f90: Fix dg directive.
         * pr16433.f: Likewise.
         * use_6.f90: Likewise.
         * use_7.f90: Likewise.
         * entry_9.f90: Likewise.
         * pointer_intent_2.f90: Likewise.
         * protected_1.f90: Likewise.
         * fmt_missing_period_2.f: Likewise.
         * pointer_intent_3.f90: Likewise.
         * protected_2.f90: Likewise.
         * protected_3.f90: Likewise.
         * pr15754.f90: Likewise.
         * protected_4.f90: Likewise.
         * protected_5.f90: Likewise.
         * initialization_2.f90: Likewise.
         * protected_6.f90: Likewise.
         * continuation_7.f90: Likewise.
         * derived_function_interface_1.f90: Likewise.
         * der_pointer_4.f90: Likewise.
         * fmt_zero_precision.f90: Likewise.
         * blockdata_3.f90: Likewise.
         * fmt_missing_period_1.f: Likewise.
         * real_const_3.f90: Likewise.
         * use_4.f90: Likewise.
         * fmt_missing_period_3.f: Likewise.
         * use_5.f90: Likewise.

More information about the Gcc-patches mailing list