In-Reply-To: <4A7144A6.5060704@gnu.org> (Paolo Bonzini's message of "Thu\,
	30 Jul 2009 08\:58\:46 +0200")
References: <cover.1248918836.git.quintela@redhat.com>
	<ab3233d773b11a0c044a270d09e6e5ad261c13f9.1248918837.git.quintela@redhat.com>
	<4A7144A6.5060704@gnu.org>
From: Juan Quintela <quintela@redhat.com>
Date: Thu, 30 Jul 2009 09:16:59 +0200
Message-ID: <m38wi61vys.fsf@neno.mitica>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Subject: [Qemu-devel] Re: [PATCH 30/42] Rename CPPFLAGS to CFLAGS
To: Paolo Bonzini <bonzini@gnu.org>
Cc: aliguori@us.ibm.com, qemu-devel@nongnu.org

Paolo Bonzini <bonzini@gnu.org> wrote:
> Just wanted to point out that this is wrong.  -I and -D *are* CPPFLAGS.

Tell that to sdl-config --cflags and friends :p

> However, the real issue is that CFLAGS should be reserved for the user
> at make time (defaulting to something *constant* like -O2 -g or
> whatever else).  Actually so should CPPFLAGS, but at least it is much
> less likely to be used, while "make CFLAGS=-g" is common during
> development and supported by almost every package I know of.
>
> You can also call it QEMU_CPPFLAGS or QEMU_CFLAGS and we can be happy.
> You then have to add both $(QEMU_CPPFLAGS) and $(CFLAGS) to
> compilation commands in both configure and Make.

If there is an agreement about that, I don't care remaining.  As things
stand now, the difference between CFLAGS and CPPFLAGS in qemu was
artificial.

Later, Juan.