From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from pasmtpa.tele.dk ([80.160.77.114]:53396 "EHLO pasmtpA.tele.dk" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1755215AbYFHVMu (ORCPT ); Sun, 8 Jun 2008 17:12:50 -0400 Date: Sun, 8 Jun 2008 23:13:31 +0200 From: Sam Ravnborg Subject: Re: [PATCH] Speed up "make headers_*" Message-ID: <20080608211331.GA32741@uranus.ravnborg.org> References: <20080608094730.GA30098@uranus.ravnborg.org> <20080608200727.GA32141@uranus.ravnborg.org> <1212958734.2534.51.camel@shinybook.infradead.org> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <1212958734.2534.51.camel@shinybook.infradead.org> Sender: linux-kbuild-owner@vger.kernel.org List-ID: To: David Woodhouse Cc: Mike Frysinger , linux-kbuild , LKML , Linus Torvalds On Sun, Jun 08, 2008 at 09:58:54PM +0100, David Woodhouse wrote: > It doesn't look like headers_check_all will work after your patches -- > but to be fair I don't think it worked _before_ them either. When you > run headers_install_all, there's no asm/ directory -- and thus the check > will fail for any headers in linux/ which try to include . > > Did this _ever_ work? It relied on what was installed in asm previously resulting in arbitary failures - which confused me a little. What I did in headers_check.pl was to introduce a second try. If we see a #include then we try both of: asm/file.h (first) asm-$arch/file.h (if the asm/file.h did not succeed) So with my version it actually works as expected and we hit a bug in cris. I've also implmented the feature that we stop on error so when cris has a bug we do not try the rest. It is both good and bad. Sam