From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:4830:134:3::10]:51932) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1WmUH5-0007IE-JS for qemu-devel@nongnu.org; Mon, 19 May 2014 16:35:13 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1WmUGz-0000dr-Ct for qemu-devel@nongnu.org; Mon, 19 May 2014 16:35:07 -0400 Received: from cantor2.suse.de ([195.135.220.15]:51255 helo=mx2.suse.de) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1WmUGz-0000cG-7H for qemu-devel@nongnu.org; Mon, 19 May 2014 16:35:01 -0400 Message-ID: <537A6AF2.4030307@suse.de> Date: Mon, 19 May 2014 22:34:58 +0200 From: Alexander Graf MIME-Version: 1.0 References: <1400367823-32610-1-git-send-email-marc.mari.barcelo@gmail.com> <1400367823-32610-4-git-send-email-marc.mari.barcelo@gmail.com> <20140519184526.00364e81@crunchbang> <537A363E.309@twiddle.net> In-Reply-To: Content-Type: text/plain; charset=UTF-8; format=flowed Content-Transfer-Encoding: quoted-printable Subject: Re: [Qemu-devel] [PATCH v3 03/16] s390: Convert debug printfs to QEMU_DPRINTF List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Peter Crosthwaite , Richard Henderson Cc: =?UTF-8?B?TWFyYyBNYXLDrQ==?= , Stefan Hajnoczi , =?UTF-8?B?QW5kcmVhcyBGw6RyYmVy?= , Overall , "qemu-devel@nongnu.org Developers" On 19.05.14 22:33, Peter Crosthwaite wrote: > On Tue, May 20, 2014 at 2:50 AM, Richard Henderson wr= ote: >> On 05/19/2014 09:45 AM, Marc Mar=C3=AD wrote: >>>>> This STDOUT vs _log() choice is a bit irregular, and I think you mi= ght >>>>> be better off abandoning it completely. Richard, Alex, do we really >>>>> need to optionally route printfery to log or stderr? (considering _= log >>>>> is NOW stderr by default now, and -D option gives you some flexibil= ity >>>>> there). Can we have just log and drop STDOUT mode? >>>>> >>>>> Regards, >>>>> Peter >>> QEMU_DPRINTF outputs to stderr, and, as you say qemu_log does it too. >>> Should QEMU_DPRINTF be removed and leave only qemu_log? > Ahh I see now, you want to use the common factored-out code in P1 > which is stderr specific. Perhaps that should use qemu_log. One option > would be to create a variant of it in common code that does uses > qemu_log instead of stderr and use that variant here. That way you can > preserve existing behaviour (and maintainer expectations), as much as > possible. I think it's safe to assume that all TCG code wants to use qemu_log. Alex