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: [DOC] clearer description of -std=gnuX


On 17/01/2008, Joseph S. Myers <joseph@codesourcery.com> wrote:
> On Thu, 17 Jan 2008, Manuel López-Ibáñez wrote:
>
> Neither is a subset of the other.  Some programs are valid in -std=c89 but
> not -std=gnu89 (e.g. programs using asm as an identifier).  Some are valid
> in -std=gnu89 but not -std=c89 (e.g. programs using the asm extension in a
> way that is not valid syntax if asm is an identifier).  Some are valid in
> both but with different semantics (e.g. programs with trigraphs in a
> string).
>

OK. I didn't get it completely right (didn't know about the
trigraphs). But I still think the description in the manual is
confusing.

Can't we agree on a better wording?

Some proposals:

-std=X        "Turns off certain GNU extensions that are incompatible with X"
-std=gnuX   "Turns on all GNU extensions based on X".

-std=X        "Accepts X programs and GNU extensions that don't
conflict with X."
-std=gnuX  "Accepts X programs that use GNU extensions".

-std=X        "Support all X programs (GNU extensions that conflict
with X are disabled)"
-std=gnuX   "Support X programs that don't conflict with GNU extensions".

Any other is welcome.

Cheers,

Manuel.


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