From mboxrd@z Thu Jan 1 00:00:00 1970 From: Chris Wilson Subject: Re: [PATCH 0/5] mm, highmem: kmap_atomic rework -v3 Date: Wed, 27 Oct 2010 11:27:57 +0100 Message-ID: <849307$a582r7@azsmga001.ch.intel.com> References: <20100918155326.478277313@chello.nl> Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=ISO-8859-1 Content-Transfer-Encoding: QUOTED-PRINTABLE Return-path: Received: from mga14.intel.com ([143.182.124.37]:12717 "EHLO mga14.intel.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1750889Ab0J0K2D (ORCPT ); Wed, 27 Oct 2010 06:28:03 -0400 In-Reply-To: <20100918155326.478277313@chello.nl> Sender: linux-arch-owner@vger.kernel.org List-ID: To: Peter Zijlstra , Linus Torvalds Cc: Andrew Morton , linux-kernel@vger.kernel.org, linux-mm@kvack.org, linux-arch@vger.kernel.org On Sat, 18 Sep 2010 17:53:26 +0200, Peter Zijlstra wrote: > Next version of the kmap_atomic rework using Andrew's fancy CPP trick= ery to > avoid having to convert the whole tree at once. >=20 > This is compile tested for i386-allmodconfig, frv, mips-sb1250-swarm, > powerpc-ppc6xx_defconfig, sparc32_defconfig, arm-omap3 (all with > HIGHEM=3Dy). This break on x86, HIGHMEM=3Dn: arch/x86/mm/iomap_32.c: In function =E2=80=98kmap_atomic_prot_pfn=E2=80= =99: arch/x86/mm/iomap_32.c:64: error: implicit declaration of function =E2=80=98kmap_atomic_idx_push=E2=80=99 arch/x86/mm/iomap_32.c: In function =E2=80=98iounmap_atomic=E2=80=99: arch/x86/mm/iomap_32.c:101: error: implicit declaration of function =E2=80=98kmap_atomic_idx_pop=E2=80=99 The use of the kmap idx there looks a little delicate so I'm not sure h= ow to fix this. -Chris --=20 Chris Wilson, Intel Open Source Technology Centre From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mga14.intel.com ([143.182.124.37]:12717 "EHLO mga14.intel.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1750889Ab0J0K2D (ORCPT ); Wed, 27 Oct 2010 06:28:03 -0400 Message-ID: <849307$a582r7@azsmga001.ch.intel.com> Date: Wed, 27 Oct 2010 11:27:57 +0100 Subject: Re: [PATCH 0/5] mm, highmem: kmap_atomic rework -v3 References: <20100918155326.478277313@chello.nl> From: Chris Wilson In-Reply-To: <20100918155326.478277313@chello.nl> Sender: linux-arch-owner@vger.kernel.org List-ID: To: Peter Zijlstra , Linus Torvalds Cc: Andrew Morton , linux-kernel@vger.kernel.org, linux-mm@kvack.org, linux-arch@vger.kernel.org Message-ID: <20101027102757.xuUEW-w_zyJuzNV4svWfNGLINkF6qcbWUt1NusVvjXg@z> On Sat, 18 Sep 2010 17:53:26 +0200, Peter Zijlstra wrote: > Next version of the kmap_atomic rework using Andrew's fancy CPP trickery to > avoid having to convert the whole tree at once. > > This is compile tested for i386-allmodconfig, frv, mips-sb1250-swarm, > powerpc-ppc6xx_defconfig, sparc32_defconfig, arm-omap3 (all with > HIGHEM=y). This break on x86, HIGHMEM=n: arch/x86/mm/iomap_32.c: In function ‘kmap_atomic_prot_pfn’: arch/x86/mm/iomap_32.c:64: error: implicit declaration of function ‘kmap_atomic_idx_push’ arch/x86/mm/iomap_32.c: In function ‘iounmap_atomic’: arch/x86/mm/iomap_32.c:101: error: implicit declaration of function ‘kmap_atomic_idx_pop’ The use of the kmap idx there looks a little delicate so I'm not sure how to fix this. -Chris -- Chris Wilson, Intel Open Source Technology Centre