From mboxrd@z Thu Jan 1 00:00:00 1970 From: Hitoshi Mitake Subject: Re: [PATCH 1/1] edac x38: new MC driver module Date: Tue, 25 Nov 2008 11:55:10 +0900 Message-ID: <20081125115510.108fa886.h.mitake@gmail.com> References: <20081105222911.d76e7e1c.mitake@clustcom.com> <413709.12821.qm@web50106.mail.re2.yahoo.com> <20081106164641.ed369060.akpm@linux-foundation.org> <20081107152830.a42766f3.mitake@clustcom.com> <20081106223122.8a255211.akpm@linux-foundation.org> <20081107153824.0ec934e6.mitake@clustcom.com> <20081106231102.aab83cd4.akpm@linux-foundation.org> <20081109112646.97c594b5.akpm@linux-foundation.org> <20081118121620.GB8625@linux-mips.org> <20081118123215.GB30509@flint.arm.linux.org.uk> <20081121011941.26e05a25.h.mitake@gmail.com> <4929ECB7.8080108@zytor.com> <57C9024A16AD2D4C97DC78E552063EA35C517F42@orsmsx505.amr.corp.intel.com> <492AEC40.5060009@zytor.com> Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Return-path: Received: from wa-out-1112.google.com ([209.85.146.178]:10551 "EHLO wa-out-1112.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752751AbYKYCzQ (ORCPT ); Mon, 24 Nov 2008 21:55:16 -0500 Received: by wa-out-1112.google.com with SMTP id v27so1126609wah.21 for ; Mon, 24 Nov 2008 18:55:15 -0800 (PST) In-Reply-To: <492AEC40.5060009@zytor.com> Sender: linux-arch-owner@vger.kernel.org List-ID: To: "H. Peter Anvin" Cc: "Luck, Tony" , Russell King , Ralf Baechle , Andrew Morton , Doug Thompson , "dougthompson@xmission.com" , "linux-kernel@vger.kernel.org" , "linux-arch@vger.kernel.org" On Mon, 24 Nov 2008 10:02:40 -0800 "H. Peter Anvin" wrote: Thanks for your reviews and advices, Peter and Tony. > Luck, Tony wrote: > >> volatile u32 __iomem *__p = addr; > >> u32 __l, __h; > > > > Do we really need all the "__" inside an inline function? Why > > not just call these "p", "l" and "h"? > > Sorry, user space habit (in userspace, and in anything that can be > included from userspace, one needs the __ to keep the namespace clean.) I think which need __ is variable type name, not each variable name. Name of each local variables can't effect namespace. And I found this comment in asm-generic/int-ll64.h, /* * __xx is ok: it doesn't pollute the POSIX namespace. Use these in the * header files exported to user space */ According to your advice, I rewrote the patch, how is this? description of this patch: Adding config value to x86 architecture to determine existence of readq/writeq Signed-off-by: Hitoshi Mitake --- arch/x86/Kconfig | 3 +++ arch/x86/include/asm/io.h | 17 +++++++++++++++++ 2 files changed, 20 insertions(+), 0 deletions(-) diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig index ac22bb7..8f3c949 100644 --- a/arch/x86/Kconfig +++ b/arch/x86/Kconfig @@ -11,9 +11,12 @@ config 64BIT config X86_32 def_bool !64BIT + select ARCH_HAS_READQ config X86_64 def_bool 64BIT + select ARCH_HAS_READQ + select ARCH_HAS_WRITEQ ### Arch settings config X86 diff --git a/arch/x86/include/asm/io.h b/arch/x86/include/asm/io.h index ac2abc8..5322d18 100644 --- a/arch/x86/include/asm/io.h +++ b/arch/x86/include/asm/io.h @@ -4,6 +4,7 @@ #define ARCH_HAS_IOREMAP_WC #include +#include #define build_mmio_read(name, size, type, reg, barrier) \ static inline type name(const volatile void __iomem *addr) \ @@ -57,6 +58,22 @@ build_mmio_write(__writeq, "q", unsigned long, "r", ) /* Let people know we have them */ #define readq readq #define writeq writeq + +#else /* CONFIG_X86_32 from here */ + +static inline __u64 readq(const volatile void __iomem *addr) +{ + const volatile u32 __iomem *p = addr; + u32 l, h; + + l = readl(p); + h = readl(p+1); + + return l + ((u64)h << 32); +} + +#define readq readq + #endif extern int iommu_bio_merge; -- 1.5.6.5