From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S964819AbWIDMg2 (ORCPT ); Mon, 4 Sep 2006 08:36:28 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S964803AbWIDMg2 (ORCPT ); Mon, 4 Sep 2006 08:36:28 -0400 Received: from mx1.redhat.com ([66.187.233.31]:26585 "EHLO mx1.redhat.com") by vger.kernel.org with ESMTP id S964819AbWIDMg1 (ORCPT ); Mon, 4 Sep 2006 08:36:27 -0400 From: David Howells In-Reply-To: <20060904115845.GP4416@stusta.de> References: <20060904115845.GP4416@stusta.de> <20060903220657.GG4416@stusta.de> <14367.1157361985@warthog.cambridge.redhat.com> To: Adrian Bunk Cc: David Howells , linux-kernel@vger.kernel.org Subject: Re: frv compile error in set_pte() X-Mailer: MH-E 8.0; nmh 1.1; GNU Emacs 22.0.50 Date: Mon, 04 Sep 2006 13:35:23 +0100 Message-ID: <30253.1157373323@warthog.cambridge.redhat.com> Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Adrian Bunk wrote: > > Does your compiler support the 'M' and 'U' constraint modifiers on FRV? > > It's a gcc 4.1.1 from ftp.gnu.org. I presume this: #define set_pte(pteptr, pteval) \ do { \ *(pteptr) = (pteval); \ asm volatile("dcf %M0" :: "U"(*pteptr)); \ } while(0) void jump(unsigned long *ppte, unsigned long pte) { set_pte(ppte, pte); } Shows the problem. I was in the process of checking that all the FRV constraint stuff got upstream, but was stalled by an ICE whilst building the gcc-4.2. This logged as gcc bug #28583 and not as yet fixed. However, I've run the above test program with the xgcc intermediate compiler built during the compilation of gcc, and that shows the problem you're reporting. I'll chase it up. Meanwhile, if you want to use a working, though older gcc, you can find one in: ftp://ftp.ges.redhat.com/private/releng/frv-060512-Fc6734/tools.tar.bz2 David -- VGER BF report: H 0.0672117