From mboxrd@z Thu Jan 1 00:00:00 1970 From: Peter Hurley Subject: Re: bitfield structures Date: Thu, 16 Oct 2014 11:57:18 -0400 Message-ID: <543FEADE.1010407@hurleysoftware.com> References: Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Return-path: Received: from n23.mail01.mtsvc.net (mailout32.mail01.mtsvc.net [216.70.64.70]) by gabe.freedesktop.org (Postfix) with ESMTP id 1D01D6E291 for ; Thu, 16 Oct 2014 08:57:21 -0700 (PDT) In-Reply-To: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: dri-devel-bounces@lists.freedesktop.org Sender: "dri-devel" To: Alex Deucher , Maling list - DRI developers List-Id: dri-devel@lists.freedesktop.org On 10/16/2014 10:14 AM, Alex Deucher wrote: > Are there any strong objections to these sorts of structures? You may want to blacklist certain compiler version/arch combinations, or get the affected arches to do it. gcc up to 4.7.1 on ia64 and ppc64 generates 64-bit wide RMW cycles on bitfields, regardless of the specified type or actual field width. The enlarged write overwrites adjacent fields. https://gcc.gnu.org/bugzilla/show_bug.cgi?id=52080 Regards, Peter Hurley