[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [PATCH 1/3] configure: Enforce gnu99 C language standard
From: |
Mike Gilbert |
Subject: |
Re: [PATCH 1/3] configure: Enforce gnu99 C language standard |
Date: |
Thu, 2 Apr 2020 18:03:46 -0400 |
On Thu, Apr 2, 2020 at 1:38 PM Leif Lindholm <address@hidden> wrote:
>
> On Thu, Apr 02, 2020 at 19:08:21 +0200, Daniel Kiper wrote:
> > > > -# Optimization flag. Allow user to override.
> > > > +if test "x$BUILD_CFLAGS" = x; then
> > > > + BUILD_CFLAGS='-std=gnu99'
> > > > +fi
> > > > +
> > > > +if test "x$HOST_CFLAGS" = x; then
> > > > + HOST_CFLAGS='-std=gnu99'
> > > > +fi
> > > > +
> > > > if test "x$TARGET_CFLAGS" = x; then
> > > > - TARGET_CFLAGS="$TARGET_CFLAGS -Os"
> > > > + TARGET_CFLAGS='-Os -std=gnu99'
> > > > fi
> > >
> > > Do these not add -std=gnu99 only if *_CFLAGS are not specified?
> >
> > Exactly...
> >
> > > Would we not want to override these always?
> > >
> > > For example, Debian/Ubuntu override HOST_CFLAGS when building grub.
> >
> > I was not sure about it. So, after some thinking I decided to give
> > a user a chance to override C language type using *_CFLAGS. However,
> > I am not so strongly tied to that. If you think we should add
> > "-std=gnu99" unconditionally I am OK with that.
>
> I think we should. As long as we permit overriding *_CFLAGS, and put
> the user-provided flags last, that would still permit someone to
> override the --std option witn an alternative one. (Works just like
> the -O flags.)
Agreed.
[PATCH 3/3] autogen: Replace -iname with -ipath in find command, Daniel Kiper, 2020/04/02