From mboxrd@z Thu Jan  1 00:00:00 1970
Received: from eggs.gnu.org ([140.186.70.92]:36631)
	by lists.gnu.org with esmtp (Exim 4.71)
	(envelope-from <gceq-qemu-devel@m.gmane.org>) id 1RLZjf-0002Qz-Ik
	for qemu-devel@nongnu.org; Wed, 02 Nov 2011 08:16:04 -0400
Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71)
	(envelope-from <gceq-qemu-devel@m.gmane.org>) id 1RLZje-0002dV-Hc
	for qemu-devel@nongnu.org; Wed, 02 Nov 2011 08:16:03 -0400
Received: from lo.gmane.org ([80.91.229.12]:42423)
	by eggs.gnu.org with esmtp (Exim 4.71)
	(envelope-from <gceq-qemu-devel@m.gmane.org>) id 1RLZje-0002dQ-CT
	for qemu-devel@nongnu.org; Wed, 02 Nov 2011 08:16:02 -0400
Received: from list by lo.gmane.org with local (Exim 4.69)
	(envelope-from <gceq-qemu-devel@m.gmane.org>) id 1RLZjc-0006SE-Uy
	for qemu-devel@nongnu.org; Wed, 02 Nov 2011 13:16:00 +0100
Received: from 93-34-200-62.ip51.fastwebnet.it ([93.34.200.62])
	by main.gmane.org with esmtp (Gmexim 0.1 (Debian))
	id 1AlnuQ-0007hv-00
	for <qemu-devel@nongnu.org>; Wed, 02 Nov 2011 13:16:00 +0100
Received: from pbonzini by 93-34-200-62.ip51.fastwebnet.it with local (Gmexim
	0.1 (Debian)) id 1AlnuQ-0007hv-00
	for <qemu-devel@nongnu.org>; Wed, 02 Nov 2011 13:16:00 +0100
From: Paolo Bonzini <pbonzini@redhat.com>
Date: Wed, 02 Nov 2011 13:15:46 +0100
Message-ID: <j8rc9j$d4$1@dough.gmane.org>
References: <1320175230-27980-1-git-send-email-ehabkost@redhat.com>
	<1320175230-27980-6-git-send-email-ehabkost@redhat.com>
	<j8qrng$f1c$1@dough.gmane.org>
	<20111102115641.GE29718@otherpad.lan.raisama.net>
Mime-Version: 1.0
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Content-Transfer-Encoding: 7bit
In-Reply-To: <20111102115641.GE29718@otherpad.lan.raisama.net>
Subject: Re: [Qemu-devel] [RFC PATCH 05/11] qemu_fclose: return last_error
	if set
List-Id: <qemu-devel.nongnu.org>
List-Unsubscribe: <https://lists.nongnu.org/mailman/options/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: <https://lists.nongnu.org/mailman/listinfo/qemu-devel>,
	<mailto:qemu-devel-request@nongnu.org?subject=subscribe>
To: qemu-devel@nongnu.org

On 11/02/2011 12:56 PM, Eduardo Habkost wrote:
> No, if it's positive it won't be set on last_error, so we have to save
> it somewhere other than last_error (that's what the qemu_close() return
> value is used for).

Ok, I was confused by your patch 6, which basically removes the only 
case when qemu_fclose was returning a positive, nonzero value. :)  I 
guess the problem is there?

> Without a separate function and qemu_file_set_if_error(), the function
> will look like:
>
> int qemu_fclose(QEMUFile *f)
> {
>      int ret = 0;
>      qemu_fflush();
>      if (f->close) {
>          ret = f->close(f->opaque);
>      }
>      if (f->last_error) {
            ^^^^^^^^^^^^^

"if (ret >= 0 && f->last_error)" perhaps?

>          ret = f->last_error;
>      }
>      g_free(f);
>      return ret;
> }
>
>
> Now that I am looking at the resulting code, it doesn't look too bad. I
> guess I was simply too eager to encapsulate every bit of logic (in this
> case the "if (f->close) ..." part) into separate functions. I find the
> two-function version slightly easier to analyze, though.

Yes, it's the same for me too now that I actually understand what's 
going on.

Paolo