Messages in this thread | | | Date | Mon, 19 Mar 2012 15:12:37 -0300 | From | Arnaldo Carvalho de Melo <> | Subject | Re: [PATCH] perf report: Add a simple GTK2-based 'perf report' browser |
| |
Em Fri, Feb 24, 2012 at 12:05:56PM +0200, Pekka Enberg escreveu: > On Fri, Feb 24, 2012 at 11:48 AM, Ingo Molnar <mingo@elte.hu> wrote: > > * Pekka Enberg <penberg@kernel.org> wrote: > > > >> On Thu, Feb 23, 2012 at 10:33 PM, Colin Walters <walters@verbum.org> wrote: > >> >> Sure. We don't want to do that for all files. Just for the ones that > >> >> include <gtk/gtk.h>. > >> > > >> > #pragma GCC diagnostic push > >> > #pragma GCC diagnostic ignored "-Wstrict-prototypes" > >> > #include <gtk/gtk.h> > >> > #pragma GCC diagnostic pop > >> > >> It's cleaner to do it at Makefile level. We should do > >> something like sparse.git Makefile does where you can > >> optionally specify CFLAGS for individual source files. > > > > I actually like the #pragma hack because it only turns off the > > check for that broken header and keeps our checks in place for > > the rest of the .c file. > > > > Could be turned into a util/gtk.h file that is included instead > > of <gtk/gtk.h>, so that we don't have to see the #pragma > > workaround all the time? > > Sure, makes sense.
Using just:
#pragma GCC diagnostic ignored "-Wstrict-prototypes" #include <gtk/gtk.h> #pragma GCC diagnostic error "-Wstrict-prototypes"
Since push/pop was introduced in gcc 4.6, and here at the test machine using RHEL6.2 I have gcc 4.4.6.
- Arnaldo
| |