From mboxrd@z Thu Jan  1 00:00:00 1970
Received: from [140.186.70.92] (port=42136 helo=eggs.gnu.org)
	by lists.gnu.org with esmtp (Exim 4.43) id 1P0AWT-0003Wj-Mo
	for qemu-devel@nongnu.org; Mon, 27 Sep 2010 06:01:30 -0400
Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.69)
	(envelope-from <paolo.bonzini@gmail.com>) id 1P0ANJ-00014W-1H
	for qemu-devel@nongnu.org; Mon, 27 Sep 2010 05:51:58 -0400
Received: from mail-iw0-f173.google.com ([209.85.214.173]:64879)
	by eggs.gnu.org with esmtp (Exim 4.69)
	(envelope-from <paolo.bonzini@gmail.com>) id 1P0ANI-00014N-SB
	for qemu-devel@nongnu.org; Mon, 27 Sep 2010 05:51:57 -0400
Received: by iwn38 with SMTP id 38so5183876iwn.4
	for <qemu-devel@nongnu.org>; Mon, 27 Sep 2010 02:51:56 -0700 (PDT)
Sender: Paolo Bonzini <paolo.bonzini@gmail.com>
Message-ID: <4CA06938.7030106@redhat.com>
Date: Mon, 27 Sep 2010 11:51:52 +0200
From: Paolo Bonzini <pbonzini@redhat.com>
MIME-Version: 1.0
References: <AANLkTik--YoLr8OqDeYEcaxjmbsC6PM2aNbcXSKeHZE1@mail.gmail.com>
	<m3bp7j8u3n.fsf@blackfin.pond.sub.org>
In-Reply-To: <m3bp7j8u3n.fsf@blackfin.pond.sub.org>
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: 7bit
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: Markus Armbruster <armbru@redhat.com>
Cc: Blue Swirl <blauwirbel@gmail.com>, qemu-devel <qemu-devel@nongnu.org>

On 09/27/2010 10:32 AM, Markus Armbruster wrote:
> Why the conditional?  cmp -s fails fine when argument files don't exist.
> 
> Note that common versions of make including GNU Make do not stat a
> rule's target to check whether the rule changed it.  Instead, they
> assume it changed, and remake everything depending on it.
> 
>      armbru@blackfin:~/tmp$ cat Makefile
>      foo: bar
>              echo "Remaking foo"
> 
>      bar:
>              [ -f $@ ] || touch $@&&  echo "Touched bar"
>      armbru@blackfin:~/tmp$ rm -f foo
>      armbru@blackfin:~/tmp$ make
>      [ -f bar ] || touch bar&&  echo "Touched bar"
>      Touched bar
>      echo "Remaking foo"
>      Remaking foo
>      armbru@blackfin:~/tmp$ make
>      echo "Remaking foo"
>      Remaking 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
	$(call quiet-command,sh $(SRC_PATH)/tracetool --$(TRACE_BACKEND) -h < $< > $@,"  GEN  trace.h)
	@cmp $@ trace.h >/dev/null 2>&1 || cp $@ trace.h

(untested).

Paolo