This is the mail archive of the gcc-patches@gcc.gnu.org mailing list for the GCC project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: [PATCH] PR28901 Add two levels for -Wunused-const-variable.


On Mon, Feb 22, 2016 at 2:43 PM, Mark Wielaard <mjw@redhat.com> wrote:
> On Mon, Feb 22, 2016 at 11:57:56AM -0700, Jeff Law wrote:
>> On 02/20/2016 06:42 PM, Mark Wielaard wrote:
>> Note that given the discussion in the BZ, I'm going to consider this a
>> regression and thus eligible for the trunk.
>
> Thanks. Unfortunately new warnings always seem to make some people
> unhappy (even when others are happy and see them as useful). Hopefully
> this compromise makes it so that nobody sees this warning as regression.
>
>> >diff --git a/gcc/cgraphunit.c b/gcc/cgraphunit.c
>> >index 0a745f0..27a073a 100644
>> >--- a/gcc/cgraphunit.c
>> >+++ b/gcc/cgraphunit.c
>> >@@ -971,7 +974,7 @@ check_global_declaration (symtab_node *snode)
>> >             (TREE_CODE (decl) == FUNCTION_DECL)
>> >             ? OPT_Wunused_function
>> >             : (TREE_READONLY (decl)
>> >-               ? OPT_Wunused_const_variable
>> >+               ? OPT_Wunused_const_variable_
>> Typo here?
>>
>> If that's not a typo, then just say so and this is approved.
>
> As Jakub already explained that was deliberate. It is how a warning
> option that can take a level is represented. Pushed.
>

It caused:

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=69911


-- 
H.J.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]