From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:4830:134:3::10]:43369) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1d2Zb9-0000ym-LC for qemu-devel@nongnu.org; Mon, 24 Apr 2017 04:43:56 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1d2Zb6-0001dS-J0 for qemu-devel@nongnu.org; Mon, 24 Apr 2017 04:43:55 -0400 Received: from mx1.redhat.com ([209.132.183.28]:41180) by eggs.gnu.org with esmtps (TLS1.0:DHE_RSA_AES_256_CBC_SHA1:32) (Exim 4.71) (envelope-from ) id 1d2Zb6-0001dG-CT for qemu-devel@nongnu.org; Mon, 24 Apr 2017 04:43:52 -0400 Received: from smtp.corp.redhat.com (int-mx05.intmail.prod.int.phx2.redhat.com [10.5.11.15]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by mx1.redhat.com (Postfix) with ESMTPS id 2AB094E4F3 for ; Mon, 24 Apr 2017 08:43:51 +0000 (UTC) Date: Mon, 24 Apr 2017 16:43:48 +0800 From: Fam Zheng Message-ID: <20170424084348.GG316@lemon.lan> References: <20170421122710.15373-1-famz@redhat.com> <20170421122710.15373-7-famz@redhat.com> <20170424082827.GC20809@redhat.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20170424082827.GC20809@redhat.com> Subject: Re: [Qemu-devel] [PATCH 06/14] crypto: Make errp the last parameter of functions List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: "Daniel P. Berrange" Cc: qemu-devel@nongnu.org, Markus Armbruster On Mon, 04/24 09:28, Daniel P. Berrange wrote: > On Fri, Apr 21, 2017 at 08:27:02PM +0800, Fam Zheng wrote: > > Move opaque to 2nd instead of the 2nd to last, so that compilers help > > check with the convertion. > > Moving 'opaque' like this should not be done. > > If you want the compiler to check the fixes, it should be done in > just two stages. First move errp & move opaque to start, compile > it & verify. Then put opaque back to where it was, and compile > again. The resulting commit thus only has the errp move, not the > unrelated & uneccessary opaque move. The idea is to let everyone's compiler verifies this patch, and also to avoid possible bugs introduced in backporting/rebasing/merging - for example if a patch in another tree addes one more implementation that uses the old order, we can notice. If you don't like this, we can be careful and don't move opaque; or after a short while, move opaque back in a separate commit (since Markus already sent a pull request). Fam