From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1753709Ab2GXKGb (ORCPT ); Tue, 24 Jul 2012 06:06:31 -0400 Received: from cam-admin0.cambridge.arm.com ([217.140.96.50]:43218 "EHLO cam-admin0.cambridge.arm.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753458Ab2GXKGa (ORCPT ); Tue, 24 Jul 2012 06:06:30 -0400 Date: Tue, 24 Jul 2012 11:05:11 +0100 From: Catalin Marinas To: Cyril Chemparathy Cc: "linux-arm-kernel@lists.infradead.org" , "linux-kernel@vger.kernel.org" , "nico@linaro.org" , Will Deacon , Vitaly Andrianov Subject: Re: [RFC 02/23] ARM: LPAE: use signed arithmetic for mask definitions Message-ID: <20120724100511.GE20132@arm.com> References: <1343092165-9470-1-git-send-email-cyril@ti.com> <1343092165-9470-3-git-send-email-cyril@ti.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <1343092165-9470-3-git-send-email-cyril@ti.com> User-Agent: Mutt/1.5.20 (2009-06-14) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Tue, Jul 24, 2012 at 02:09:04AM +0100, Cyril Chemparathy wrote: > This patch applies to PAGE_MASK, PMD_MASK, and PGDIR_MASK, where forcing > unsigned long math truncates the mask at the 32-bits. This clearly does bad > things on PAE systems. > > This patch fixes this problem by defining these masks as signed quantities. > We then rely on sign extension to do the right thing. > > Signed-off-by: Cyril Chemparathy > Signed-off-by: Vitaly Andrianov > --- > arch/arm/include/asm/page.h | 7 ++++++- > arch/arm/include/asm/pgtable-3level.h | 6 +++--- > 2 files changed, 9 insertions(+), 4 deletions(-) > > diff --git a/arch/arm/include/asm/page.h b/arch/arm/include/asm/page.h > index ecf9019..1c810d2 100644 > --- a/arch/arm/include/asm/page.h > +++ b/arch/arm/include/asm/page.h > @@ -13,7 +13,12 @@ > /* PAGE_SHIFT determines the page size */ > #define PAGE_SHIFT 12 > #define PAGE_SIZE (_AC(1,UL) << PAGE_SHIFT) > -#define PAGE_MASK (~(PAGE_SIZE-1)) > + > +/* > + * We do not use PAGE_SIZE in the following because we rely on sign > + * extension to appropriately extend upper bits for PAE systems > + */ > +#define PAGE_MASK (~((1 << PAGE_SHIFT) - 1)) Would it work if we use a 1ULL here and avoid the sign trick? I'm worried about some context where this would lose the sign. We could have an #ifdef here while the PGDIR/PMD masks are in a separate file already. -- Catalin From mboxrd@z Thu Jan 1 00:00:00 1970 From: catalin.marinas@arm.com (Catalin Marinas) Date: Tue, 24 Jul 2012 11:05:11 +0100 Subject: [RFC 02/23] ARM: LPAE: use signed arithmetic for mask definitions In-Reply-To: <1343092165-9470-3-git-send-email-cyril@ti.com> References: <1343092165-9470-1-git-send-email-cyril@ti.com> <1343092165-9470-3-git-send-email-cyril@ti.com> Message-ID: <20120724100511.GE20132@arm.com> To: linux-arm-kernel@lists.infradead.org List-Id: linux-arm-kernel.lists.infradead.org On Tue, Jul 24, 2012 at 02:09:04AM +0100, Cyril Chemparathy wrote: > This patch applies to PAGE_MASK, PMD_MASK, and PGDIR_MASK, where forcing > unsigned long math truncates the mask at the 32-bits. This clearly does bad > things on PAE systems. > > This patch fixes this problem by defining these masks as signed quantities. > We then rely on sign extension to do the right thing. > > Signed-off-by: Cyril Chemparathy > Signed-off-by: Vitaly Andrianov > --- > arch/arm/include/asm/page.h | 7 ++++++- > arch/arm/include/asm/pgtable-3level.h | 6 +++--- > 2 files changed, 9 insertions(+), 4 deletions(-) > > diff --git a/arch/arm/include/asm/page.h b/arch/arm/include/asm/page.h > index ecf9019..1c810d2 100644 > --- a/arch/arm/include/asm/page.h > +++ b/arch/arm/include/asm/page.h > @@ -13,7 +13,12 @@ > /* PAGE_SHIFT determines the page size */ > #define PAGE_SHIFT 12 > #define PAGE_SIZE (_AC(1,UL) << PAGE_SHIFT) > -#define PAGE_MASK (~(PAGE_SIZE-1)) > + > +/* > + * We do not use PAGE_SIZE in the following because we rely on sign > + * extension to appropriately extend upper bits for PAE systems > + */ > +#define PAGE_MASK (~((1 << PAGE_SHIFT) - 1)) Would it work if we use a 1ULL here and avoid the sign trick? I'm worried about some context where this would lose the sign. We could have an #ifdef here while the PGDIR/PMD masks are in a separate file already. -- Catalin