From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S261458AbTIFNzf (ORCPT ); Sat, 6 Sep 2003 09:55:35 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S261465AbTIFNzf (ORCPT ); Sat, 6 Sep 2003 09:55:35 -0400 Received: from www.piratehaven.org ([204.253.162.40]:47595 "EHLO skull.piratehaven.org") by vger.kernel.org with ESMTP id S261458AbTIFNze (ORCPT ); Sat, 6 Sep 2003 09:55:34 -0400 Date: Sat, 6 Sep 2003 06:55:33 -0700 From: Dale Harris To: Benjamin Herrenschmidt Cc: linux-kernel mailing list Subject: Re: compile problems on PPC for 2.6.0-test4 Message-ID: <20030906135533.GN29466@maybe.org> Mail-Followup-To: Dale Harris , Benjamin Herrenschmidt , linux-kernel mailing list References: <20030906040904.GM29466@maybe.org> <1062835714.824.17.camel@gaston> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <1062835714.824.17.camel@gaston> User-Agent: Mutt/1.4i Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org On Sat, Sep 06, 2003 at 10:08:34AM +0200, Benjamin Herrenschmidt elucidated: > > It shoud not be failing on this though: No, it didn't fail, it was just a warning from the compiler. > > #define mdelay(n) (\ > (__builtin_constant_p(n) && (n)<=MAX_UDELAY_MS) ? udelay((n)*1000) : \ > ({unsigned long __ms=(n); while (__ms--) udelay(1000);})) > > The __builtin_constant_p(5000) && (5000)<=5 test should fail thus > turning into a loop of udelay(1000). Your gcc seems to be screwing > it up. What version are you using ? 2.95.4, but this is from Debian unstable, package version: 2.95.4-17. > > The current bk snapshot I have here doesn't have a problem at this line. BTW, I guess I wasn't compiling a straight vanilla kernel, it was off your rsync server, source.mvista.com::linuxppc-2.5. The vanilla kernel bombs drivers/ide/ppc/pmac.c on my laptop. -- Dale Harris rodmur@maybe.org /.-)