From mboxrd@z Thu Jan  1 00:00:00 1970
Received: from [140.186.70.92] (port=51918 helo=eggs.gnu.org)
	by lists.gnu.org with esmtp (Exim 4.43) id 1P0GKg-0008KZ-MC
	for qemu-devel@nongnu.org; Mon, 27 Sep 2010 12:13:39 -0400
Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.69)
	(envelope-from <blauwirbel@gmail.com>) id 1P0GKf-0007JJ-DR
	for qemu-devel@nongnu.org; Mon, 27 Sep 2010 12:13:38 -0400
Received: from mail-gy0-f173.google.com ([209.85.160.173]:62564)
	by eggs.gnu.org with esmtp (Exim 4.69)
	(envelope-from <blauwirbel@gmail.com>) id 1P0GKf-0007Ij-9Z
	for qemu-devel@nongnu.org; Mon, 27 Sep 2010 12:13:37 -0400
Received: by gya1 with SMTP id 1so2215698gya.4
	for <qemu-devel@nongnu.org>; Mon, 27 Sep 2010 09:13:33 -0700 (PDT)
MIME-Version: 1.0
In-Reply-To: <4CA06938.7030106@redhat.com>
References: <AANLkTik--YoLr8OqDeYEcaxjmbsC6PM2aNbcXSKeHZE1@mail.gmail.com>
	<m3bp7j8u3n.fsf@blackfin.pond.sub.org> <4CA06938.7030106@redhat.com>
From: Blue Swirl <blauwirbel@gmail.com>
Date: Mon, 27 Sep 2010 16:13:12 +0000
Message-ID: <AANLkTimPYni2UHVqu0D5TbY7jnAYbdYXRNN-H6bqL2tt@mail.gmail.com>
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: quoted-printable
Subject: [Qemu-devel] Re: [PATCH 2/2] trace: avoid unnecessary recompilation
	if nothing changed
List-Id: qemu-devel.nongnu.org
List-Unsubscribe: <http://lists.nongnu.org/mailman/listinfo/qemu-devel>,
	<mailto:qemu-devel-request@nongnu.org?subject=unsubscribe>
List-Archive: <http://lists.nongnu.org/archive/html/qemu-devel>
List-Post: <mailto:qemu-devel@nongnu.org>
List-Help: <mailto:qemu-devel-request@nongnu.org?subject=help>
List-Subscribe: <http://lists.nongnu.org/mailman/listinfo/qemu-devel>,
	<mailto:qemu-devel-request@nongnu.org?subject=subscribe>
To: Paolo Bonzini <pbonzini@redhat.com>
Cc: Markus Armbruster <armbru@redhat.com>, qemu-devel <qemu-devel@nongnu.org>

On Mon, Sep 27, 2010 at 9:51 AM, Paolo Bonzini <pbonzini@redhat.com> wrote:
> On 09/27/2010 10:32 AM, Markus Armbruster wrote:
>> Why the conditional? =C2=A0cmp -s fails fine when argument files don't e=
xist.
>>
>> Note that common versions of make including GNU Make do not stat a
>> rule's target to check whether the rule changed it. =C2=A0Instead, they
>> assume it changed, and remake everything depending on it.
>>
>> =C2=A0 =C2=A0 =C2=A0armbru@blackfin:~/tmp$ cat Makefile
>> =C2=A0 =C2=A0 =C2=A0foo: bar
>> =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0echo "Remaking foo"
>>
>> =C2=A0 =C2=A0 =C2=A0bar:
>> =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0[ -f $@ ] || touch $@&& =
=C2=A0echo "Touched bar"
>> =C2=A0 =C2=A0 =C2=A0armbru@blackfin:~/tmp$ rm -f foo
>> =C2=A0 =C2=A0 =C2=A0armbru@blackfin:~/tmp$ make
>> =C2=A0 =C2=A0 =C2=A0[ -f bar ] || touch bar&& =C2=A0echo "Touched bar"
>> =C2=A0 =C2=A0 =C2=A0Touched bar
>> =C2=A0 =C2=A0 =C2=A0echo "Remaking foo"
>> =C2=A0 =C2=A0 =C2=A0Remaking foo
>> =C2=A0 =C2=A0 =C2=A0armbru@blackfin:~/tmp$ make
>> =C2=A0 =C2=A0 =C2=A0echo "Remaking foo"
>> =C2=A0 =C2=A0 =C2=A0Remaking foo
>>
>> I doubt your patch avoids churn as advertized with such makes.
>
> Indeed, see how it's done for config-*.h.
>
> # Uses generic rule in rules.mak
> trace.h: trace.h-timestamp
> trace.h-timestamp: $(SRC_PATH)/trace-events config-host.mak
> =C2=A0 =C2=A0 =C2=A0 =C2=A0$(call quiet-command,sh $(SRC_PATH)/tracetool =
--$(TRACE_BACKEND) -h < $< > $@," =C2=A0GEN =C2=A0trace.h)
> =C2=A0 =C2=A0 =C2=A0 =C2=A0@cmp $@ trace.h >/dev/null 2>&1 || cp $@ trace=
.h
>
> (untested).

I just copied the rule from %/config-devices.mak. Is also that rule
then incorrect?

Perhaps there could be a macro for this, not unlike move-if-change
script in various GNU packages?