From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S966914AbXEGXBM (ORCPT ); Mon, 7 May 2007 19:01:12 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1754656AbXEGXBI (ORCPT ); Mon, 7 May 2007 19:01:08 -0400 Received: from mga09.intel.com ([134.134.136.24]:17602 "EHLO mga09.intel.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1754142AbXEGXBH (ORCPT ); Mon, 7 May 2007 19:01:07 -0400 X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="4.14,502,1170662400"; d="scan'208";a="84738975" Date: Mon, 7 May 2007 15:58:41 -0700 From: Fenghua Yu To: akpm@linux-foundation.org, suresh.b.siddha@intel.com, linux-kernel@vger.kernel.org, clameter@sgi.com Cc: fenghua.yu@intel.com Subject: Re: [PATCH 1/2] Define percpu smp cacheline align interface Message-ID: <20070507225841.GA13872@linux-os.sc.intel.com> References: <33E1C72C74DBE747B7B59C1740F7443701A2F0AB@orsmsx417.amr.corp.intel.com> <20070505001218.GA27093@linux-os.sc.intel.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20070505001218.GA27093@linux-os.sc.intel.com> User-Agent: Mutt/1.4.1i Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org On Fri, May 04, 2007 at 05:12:18PM -0700, Fenghua Yu wrote: > > Define percpu smp cacheline align interface This is updated patch. Per cpu cacheline aligned data section name and macro name are changed to have better code understanding. The patches place all of smp cacheline aligned percpu data into .data.percpu.shared_cacheline_aligned. Other percpu data is still in data.percpu section. The patches can reduce cache line access in SMP and reduce alignment gap waste. The patches also define PERCPU macro for vmlinux.lds.S for concise code. PATCH 1/2: Define percpu smp cacheline align interface PATCH 2/2: Call percpu smp cacheline algin interface Signed-off-by: Fenghua Yu Acked-by: Suresh Siddha --- arch/alpha/kernel/vmlinux.lds.S | 5 +---- arch/arm/kernel/vmlinux.lds.S | 5 +---- arch/cris/arch-v32/vmlinux.lds.S | 5 +---- arch/frv/kernel/vmlinux.lds.S | 5 +---- arch/i386/kernel/vmlinux.lds.S | 7 +------ arch/ia64/kernel/vmlinux.lds.S | 1 + arch/m32r/kernel/vmlinux.lds.S | 5 +---- arch/mips/kernel/vmlinux.lds.S | 5 +---- arch/parisc/kernel/vmlinux.lds.S | 7 +++---- arch/powerpc/kernel/vmlinux.lds.S | 7 +------ arch/ppc/kernel/vmlinux.lds.S | 5 +---- arch/s390/kernel/vmlinux.lds.S | 5 +---- arch/sh/kernel/vmlinux.lds.S | 5 +---- arch/sh64/kernel/vmlinux.lds.S | 5 +---- arch/sparc/kernel/vmlinux.lds.S | 5 +---- arch/sparc64/kernel/vmlinux.lds.S | 5 +---- arch/x86_64/kernel/vmlinux.lds.S | 6 ++---- arch/xtensa/kernel/vmlinux.lds.S | 5 +---- include/asm-generic/percpu.h | 8 ++++++++ include/asm-generic/vmlinux.lds.h | 10 ++++++++++ include/asm-i386/percpu.h | 5 +++++ include/asm-ia64/percpu.h | 10 ++++++++++ include/asm-powerpc/percpu.h | 7 +++++++ include/asm-s390/percpu.h | 7 +++++++ include/asm-sparc64/percpu.h | 7 +++++++ include/asm-x86_64/percpu.h | 7 +++++++ 26 files changed, 82 insertions(+), 72 deletions(-) diff -Nurp linux-2.6.21-rc7.0/arch/alpha/kernel/vmlinux.lds.S linux-2.6.21-rc7.1/arch/alpha/kernel/vmlinux.lds.S --- linux-2.6.21-rc7.0/arch/alpha/kernel/vmlinux.lds.S 2007-05-01 07:32:58.000000000 -0700 +++ linux-2.6.21-rc7.1/arch/alpha/kernel/vmlinux.lds.S 2007-05-02 17:00:29.000000000 -0700 @@ -69,10 +69,7 @@ SECTIONS . = ALIGN(8); SECURITY_INIT - . = ALIGN(8192); - __per_cpu_start = .; - .data.percpu : { *(.data.percpu) } - __per_cpu_end = .; + PERCPU(8192) . = ALIGN(2*8192); __init_end = .; diff -Nurp linux-2.6.21-rc7.0/arch/arm/kernel/vmlinux.lds.S linux-2.6.21-rc7.1/arch/arm/kernel/vmlinux.lds.S --- linux-2.6.21-rc7.0/arch/arm/kernel/vmlinux.lds.S 2007-05-01 07:32:58.000000000 -0700 +++ linux-2.6.21-rc7.1/arch/arm/kernel/vmlinux.lds.S 2007-05-02 17:00:29.000000000 -0700 @@ -59,10 +59,7 @@ SECTIONS usr/built-in.o(.init.ramfs) __initramfs_end = .; #endif - . = ALIGN(4096); - __per_cpu_start = .; - *(.data.percpu) - __per_cpu_end = .; + PERCPU(4096) #ifndef CONFIG_XIP_KERNEL __init_begin = _stext; *(.init.data) diff -Nurp linux-2.6.21-rc7.0/arch/cris/arch-v32/vmlinux.lds.S linux-2.6.21-rc7.1/arch/cris/arch-v32/vmlinux.lds.S --- linux-2.6.21-rc7.0/arch/cris/arch-v32/vmlinux.lds.S 2007-05-01 07:32:59.000000000 -0700 +++ linux-2.6.21-rc7.1/arch/cris/arch-v32/vmlinux.lds.S 2007-05-02 17:00:29.000000000 -0700 @@ -92,10 +92,7 @@ SECTIONS } SECURITY_INIT - . = ALIGN (8192); - __per_cpu_start = .; - .data.percpu : { *(.data.percpu) } - __per_cpu_end = .; + PERCPU(8192) #ifdef CONFIG_BLK_DEV_INITRD .init.ramfs : { diff -Nurp linux-2.6.21-rc7.0/arch/frv/kernel/vmlinux.lds.S linux-2.6.21-rc7.1/arch/frv/kernel/vmlinux.lds.S --- linux-2.6.21-rc7.0/arch/frv/kernel/vmlinux.lds.S 2007-05-01 07:32:59.000000000 -0700 +++ linux-2.6.21-rc7.1/arch/frv/kernel/vmlinux.lds.S 2007-05-02 17:00:29.000000000 -0700 @@ -57,10 +57,7 @@ SECTIONS __alt_instructions_end = .; .altinstr_replacement : { *(.altinstr_replacement) } - . = ALIGN(4096); - __per_cpu_start = .; - .data.percpu : { *(.data.percpu) } - __per_cpu_end = .; + PERCPU(4096) #ifdef CONFIG_BLK_DEV_INITRD . = ALIGN(4096); diff -Nurp linux-2.6.21-rc7.0/arch/i386/kernel/vmlinux.lds.S linux-2.6.21-rc7.1/arch/i386/kernel/vmlinux.lds.S --- linux-2.6.21-rc7.0/arch/i386/kernel/vmlinux.lds.S 2007-05-01 07:32:59.000000000 -0700 +++ linux-2.6.21-rc7.1/arch/i386/kernel/vmlinux.lds.S 2007-05-02 17:00:29.000000000 -0700 @@ -185,12 +185,7 @@ SECTIONS __initramfs_end = .; } #endif - . = ALIGN(4096); - .data.percpu : AT(ADDR(.data.percpu) - LOAD_OFFSET) { - __per_cpu_start = .; - *(.data.percpu) - __per_cpu_end = .; - } + PERCPU(4096) . = ALIGN(4096); /* freed after init ends here */ diff -Nurp linux-2.6.21-rc7.0/arch/ia64/kernel/vmlinux.lds.S linux-2.6.21-rc7.1/arch/ia64/kernel/vmlinux.lds.S --- linux-2.6.21-rc7.0/arch/ia64/kernel/vmlinux.lds.S 2007-05-01 07:32:59.000000000 -0700 +++ linux-2.6.21-rc7.1/arch/ia64/kernel/vmlinux.lds.S 2007-05-02 17:00:29.000000000 -0700 @@ -206,6 +206,7 @@ SECTIONS { __per_cpu_start = .; *(.data.percpu) + *(.data.percpu.shared_cacheline_aligned) __per_cpu_end = .; } . = __phys_per_cpu_start + PERCPU_PAGE_SIZE; /* ensure percpu data fits diff -Nurp linux-2.6.21-rc7.0/arch/m32r/kernel/vmlinux.lds.S linux-2.6.21-rc7.1/arch/m32r/kernel/vmlinux.lds.S --- linux-2.6.21-rc7.0/arch/m32r/kernel/vmlinux.lds.S 2007-05-01 07:32:59.000000000 -0700 +++ linux-2.6.21-rc7.1/arch/m32r/kernel/vmlinux.lds.S 2007-05-02 17:00:29.000000000 -0700 @@ -111,10 +111,7 @@ SECTIONS __initramfs_end = .; #endif - . = ALIGN(4096); - __per_cpu_start = .; - .data.percpu : { *(.data.percpu) } - __per_cpu_end = .; + PERCPU(4096) . = ALIGN(4096); __init_end = .; /* freed after init ends here */ diff -Nurp linux-2.6.21-rc7.0/arch/mips/kernel/vmlinux.lds.S linux-2.6.21-rc7.1/arch/mips/kernel/vmlinux.lds.S --- linux-2.6.21-rc7.0/arch/mips/kernel/vmlinux.lds.S 2007-05-01 07:32:59.000000000 -0700 +++ linux-2.6.21-rc7.1/arch/mips/kernel/vmlinux.lds.S 2007-05-02 17:00:29.000000000 -0700 @@ -121,10 +121,7 @@ SECTIONS .init.ramfs : { *(.init.ramfs) } __initramfs_end = .; #endif - . = ALIGN(_PAGE_SIZE); - __per_cpu_start = .; - .data.percpu : { *(.data.percpu) } - __per_cpu_end = .; + PERCPU(_PAGE_SIZE) . = ALIGN(_PAGE_SIZE); __init_end = .; /* freed after init ends here */ diff -Nurp linux-2.6.21-rc7.0/arch/parisc/kernel/vmlinux.lds.S linux-2.6.21-rc7.1/arch/parisc/kernel/vmlinux.lds.S --- linux-2.6.21-rc7.0/arch/parisc/kernel/vmlinux.lds.S 2007-05-01 07:32:59.000000000 -0700 +++ linux-2.6.21-rc7.1/arch/parisc/kernel/vmlinux.lds.S 2007-05-02 17:00:29.000000000 -0700 @@ -182,10 +182,9 @@ SECTIONS .init.ramfs : { *(.init.ramfs) } __initramfs_end = .; #endif - . = ALIGN(ASM_PAGE_SIZE); - __per_cpu_start = .; - .data.percpu : { *(.data.percpu) } - __per_cpu_end = .; + + PERCPU(ASM_PAGE_SIZE) + . = ALIGN(ASM_PAGE_SIZE); __init_end = .; /* freed after init ends here */ diff -Nurp linux-2.6.21-rc7.0/arch/powerpc/kernel/vmlinux.lds.S linux-2.6.21-rc7.1/arch/powerpc/kernel/vmlinux.lds.S --- linux-2.6.21-rc7.0/arch/powerpc/kernel/vmlinux.lds.S 2007-05-01 07:32:59.000000000 -0700 +++ linux-2.6.21-rc7.1/arch/powerpc/kernel/vmlinux.lds.S 2007-05-02 17:00:29.000000000 -0700 @@ -139,12 +139,7 @@ SECTIONS __initramfs_end = .; } #endif - . = ALIGN(PAGE_SIZE); - .data.percpu : { - __per_cpu_start = .; - *(.data.percpu) - __per_cpu_end = .; - } + PERCPU(PAGE_SIZE) . = ALIGN(8); .machine.desc : { diff -Nurp linux-2.6.21-rc7.0/arch/ppc/kernel/vmlinux.lds.S linux-2.6.21-rc7.1/arch/ppc/kernel/vmlinux.lds.S --- linux-2.6.21-rc7.0/arch/ppc/kernel/vmlinux.lds.S 2007-05-01 07:32:59.000000000 -0700 +++ linux-2.6.21-rc7.1/arch/ppc/kernel/vmlinux.lds.S 2007-05-02 17:00:29.000000000 -0700 @@ -131,10 +131,7 @@ SECTIONS __ftr_fixup : { *(__ftr_fixup) } __stop___ftr_fixup = .; - . = ALIGN(4096); - __per_cpu_start = .; - .data.percpu : { *(.data.percpu) } - __per_cpu_end = .; + PERCPU(4096) #ifdef CONFIG_BLK_DEV_INITRD . = ALIGN(4096); diff -Nurp linux-2.6.21-rc7.0/arch/s390/kernel/vmlinux.lds.S linux-2.6.21-rc7.1/arch/s390/kernel/vmlinux.lds.S --- linux-2.6.21-rc7.0/arch/s390/kernel/vmlinux.lds.S 2007-05-01 07:32:59.000000000 -0700 +++ linux-2.6.21-rc7.1/arch/s390/kernel/vmlinux.lds.S 2007-05-02 17:00:29.000000000 -0700 @@ -108,10 +108,7 @@ SECTIONS . = ALIGN(2); __initramfs_end = .; #endif - . = ALIGN(4096); - __per_cpu_start = .; - .data.percpu : { *(.data.percpu) } - __per_cpu_end = .; + PERCPU(4096) . = ALIGN(4096); __init_end = .; /* freed after init ends here */ diff -Nurp linux-2.6.21-rc7.0/arch/sh/kernel/vmlinux.lds.S linux-2.6.21-rc7.1/arch/sh/kernel/vmlinux.lds.S --- linux-2.6.21-rc7.0/arch/sh/kernel/vmlinux.lds.S 2007-05-01 07:32:59.000000000 -0700 +++ linux-2.6.21-rc7.1/arch/sh/kernel/vmlinux.lds.S 2007-05-02 17:00:29.000000000 -0700 @@ -61,10 +61,7 @@ SECTIONS . = ALIGN(PAGE_SIZE); __nosave_end = .; - . = ALIGN(PAGE_SIZE); - __per_cpu_start = .; - .data.percpu : { *(.data.percpu) } - __per_cpu_end = .; + PERCPU(PAGE_SIZE) .data.cacheline_aligned : { *(.data.cacheline_aligned) } _edata = .; /* End of data section */ diff -Nurp linux-2.6.21-rc7.0/arch/sh64/kernel/vmlinux.lds.S linux-2.6.21-rc7.1/arch/sh64/kernel/vmlinux.lds.S --- linux-2.6.21-rc7.0/arch/sh64/kernel/vmlinux.lds.S 2007-05-01 07:32:59.000000000 -0700 +++ linux-2.6.21-rc7.1/arch/sh64/kernel/vmlinux.lds.S 2007-05-02 17:00:29.000000000 -0700 @@ -86,10 +86,7 @@ SECTIONS . = ALIGN(PAGE_SIZE); .data.page_aligned : C_PHYS(.data.page_aligned) { *(.data.page_aligned) } - . = ALIGN(PAGE_SIZE); - __per_cpu_start = .; - .data.percpu : C_PHYS(.data.percpu) { *(.data.percpu) } - __per_cpu_end = . ; + PERCPU(PAGE_SIZE) .data.cacheline_aligned : C_PHYS(.data.cacheline_aligned) { *(.data.cacheline_aligned) } _edata = .; /* End of data section */ diff -Nurp linux-2.6.21-rc7.0/arch/sparc/kernel/vmlinux.lds.S linux-2.6.21-rc7.1/arch/sparc/kernel/vmlinux.lds.S --- linux-2.6.21-rc7.0/arch/sparc/kernel/vmlinux.lds.S 2007-05-01 07:32:59.000000000 -0700 +++ linux-2.6.21-rc7.1/arch/sparc/kernel/vmlinux.lds.S 2007-05-02 17:00:29.000000000 -0700 @@ -66,10 +66,7 @@ SECTIONS __initramfs_end = .; #endif - . = ALIGN(4096); - __per_cpu_start = .; - .data.percpu : { *(.data.percpu) } - __per_cpu_end = .; + PERCPU(4096) . = ALIGN(4096); __init_end = .; . = ALIGN(32); diff -Nurp linux-2.6.21-rc7.0/arch/sparc64/kernel/vmlinux.lds.S linux-2.6.21-rc7.1/arch/sparc64/kernel/vmlinux.lds.S --- linux-2.6.21-rc7.0/arch/sparc64/kernel/vmlinux.lds.S 2007-05-01 07:32:59.000000000 -0700 +++ linux-2.6.21-rc7.1/arch/sparc64/kernel/vmlinux.lds.S 2007-05-02 17:00:29.000000000 -0700 @@ -90,10 +90,7 @@ SECTIONS __initramfs_end = .; #endif - . = ALIGN(8192); - __per_cpu_start = .; - .data.percpu : { *(.data.percpu) } - __per_cpu_end = .; + PERCPU(8192) . = ALIGN(8192); __init_end = .; __bss_start = .; diff -Nurp linux-2.6.21-rc7.0/arch/x86_64/kernel/vmlinux.lds.S linux-2.6.21-rc7.1/arch/x86_64/kernel/vmlinux.lds.S --- linux-2.6.21-rc7.0/arch/x86_64/kernel/vmlinux.lds.S 2007-05-01 07:32:59.000000000 -0700 +++ linux-2.6.21-rc7.1/arch/x86_64/kernel/vmlinux.lds.S 2007-05-02 17:00:29.000000000 -0700 @@ -209,10 +209,8 @@ SECTIONS __initramfs_end = .; #endif - . = ALIGN(4096); - __per_cpu_start = .; - .data.percpu : AT(ADDR(.data.percpu) - LOAD_OFFSET) { *(.data.percpu) } - __per_cpu_end = .; + PERCPU(4096) + . = ALIGN(4096); __init_end = .; diff -Nurp linux-2.6.21-rc7.0/arch/xtensa/kernel/vmlinux.lds.S linux-2.6.21-rc7.1/arch/xtensa/kernel/vmlinux.lds.S --- linux-2.6.21-rc7.0/arch/xtensa/kernel/vmlinux.lds.S 2007-05-01 07:32:59.000000000 -0700 +++ linux-2.6.21-rc7.1/arch/xtensa/kernel/vmlinux.lds.S 2007-05-02 17:00:29.000000000 -0700 @@ -198,10 +198,7 @@ SECTIONS __ftr_fixup : { *(__ftr_fixup) } __stop___ftr_fixup = .; - . = ALIGN(4096); - __per_cpu_start = .; - .data.percpu : { *(.data.percpu) } - __per_cpu_end = .; + PERCPU(4096) #ifdef CONFIG_BLK_DEV_INITRD . = ALIGN(4096); diff -Nurp linux-2.6.21-rc7.0/include/asm-generic/percpu.h linux-2.6.21-rc7.1/include/asm-generic/percpu.h --- linux-2.6.21-rc7.0/include/asm-generic/percpu.h 2007-05-01 07:33:03.000000000 -0700 +++ linux-2.6.21-rc7.1/include/asm-generic/percpu.h 2007-05-04 11:15:26.000000000 -0700 @@ -14,6 +14,11 @@ extern unsigned long __per_cpu_offset[NR #define DEFINE_PER_CPU(type, name) \ __attribute__((__section__(".data.percpu"))) __typeof__(type) per_cpu__##name +#define DEFINE_PER_CPU_SHARED_CACHELINE_ALIGNED(type, name) \ + __attribute__((__section__(".data.percpu.shared_cacheline_aligned"))) \ + __typeof__(type) per_cpu__##name \ + ____cacheline_aligned_in_smp + /* var is in discarded region: offset to particular copy we want */ #define per_cpu(var, cpu) (*({ \ extern int simple_identifier_##var(void); \ @@ -34,6 +39,9 @@ do { \ #define DEFINE_PER_CPU(type, name) \ __typeof__(type) per_cpu__##name +#define DEFINE_PER_CPU_SHARED_CACHELINE_ALIGNED(type, name) \ + DEFINE_PER_CPU(type, name) + #define per_cpu(var, cpu) (*((void)(cpu), &per_cpu__##var)) #define __get_cpu_var(var) per_cpu__##var #define __raw_get_cpu_var(var) per_cpu__##var diff -Nurp linux-2.6.21-rc7.0/include/asm-generic/vmlinux.lds.h linux-2.6.21-rc7.1/include/asm-generic/vmlinux.lds.h --- linux-2.6.21-rc7.0/include/asm-generic/vmlinux.lds.h 2007-05-01 07:33:03.000000000 -0700 +++ linux-2.6.21-rc7.1/include/asm-generic/vmlinux.lds.h 2007-05-02 17:00:29.000000000 -0700 @@ -265,3 +265,13 @@ *(.initcall6s.init) \ *(.initcall7.init) \ *(.initcall7s.init) + +#define PERCPU(align) \ + . = ALIGN(align); \ + __per_cpu_start = .; \ + .data.percpu : AT(ADDR(.data.percpu) - LOAD_OFFSET) { \ + *(.data.percpu) \ + *(.data.percpu.shared_cacheline_aligned) \ + } \ + __per_cpu_end = .; + diff -Nurp linux-2.6.21-rc7.0/include/asm-i386/percpu.h linux-2.6.21-rc7.1/include/asm-i386/percpu.h --- linux-2.6.21-rc7.0/include/asm-i386/percpu.h 2007-05-01 07:33:03.000000000 -0700 +++ linux-2.6.21-rc7.1/include/asm-i386/percpu.h 2007-05-02 17:00:29.000000000 -0700 @@ -54,6 +54,11 @@ extern unsigned long __per_cpu_offset[]; #define DEFINE_PER_CPU(type, name) \ __attribute__((__section__(".data.percpu"))) __typeof__(type) per_cpu__##name +#define DEFINE_PER_CPU_SHARED_CACHELINE_ALIGNED(type, name) \ + __attribute__((__section__(".data.percpu.shared_cacheline_aligned"))) \ + __typeof__(type) per_cpu__##name \ + ____cacheline_aligned_in_smp + /* We can use this directly for local CPU (faster). */ DECLARE_PER_CPU(unsigned long, this_cpu_off); diff -Nurp linux-2.6.21-rc7.0/include/asm-ia64/percpu.h linux-2.6.21-rc7.1/include/asm-ia64/percpu.h --- linux-2.6.21-rc7.0/include/asm-ia64/percpu.h 2007-04-15 16:50:57.000000000 -0700 +++ linux-2.6.21-rc7.1/include/asm-ia64/percpu.h 2007-05-04 11:47:45.000000000 -0700 @@ -29,6 +29,16 @@ __attribute__((__section__(".data.percpu"))) \ __SMALL_ADDR_AREA __typeof__(type) per_cpu__##name +#ifdef CONFIG_SMP +#define DEFINE_PER_CPU_SHARED_CACHELINE_ALIGNED(type, name) \ + __attribute__((__section__(".data.percpu.shared_cacheline_aligned"))) \ + __SMALL_ADDR_AREA __typeof__(type) per_cpu__##name \ + ____cacheline_aligned_in_smp +#else +#define DEFINE_PER_CPU_SHARED_CACHELINE_ALIGNED(type, name) \ + DEFINE_PER_CPU(type, name) +#endif + /* * Pretty much a literal copy of asm-generic/percpu.h, except that percpu_modcopy() is an * external routine, to avoid include-hell. diff -Nurp linux-2.6.21-rc7.0/include/asm-powerpc/percpu.h linux-2.6.21-rc7.1/include/asm-powerpc/percpu.h --- linux-2.6.21-rc7.0/include/asm-powerpc/percpu.h 2007-04-15 16:50:57.000000000 -0700 +++ linux-2.6.21-rc7.1/include/asm-powerpc/percpu.h 2007-05-04 11:19:37.000000000 -0700 @@ -20,6 +20,11 @@ #define DEFINE_PER_CPU(type, name) \ __attribute__((__section__(".data.percpu"))) __typeof__(type) per_cpu__##name +#define DEFINE_PER_CPU_SHARED_CACHELINE_ALIGNED(type, name) \ + __attribute__((__section__(".data.percpu.shared_cacheline_aligned"))) \ + __typeof__(type) per_cpu__##name \ + ____cacheline_aligned_in_smp + /* var is in discarded region: offset to particular copy we want */ #define per_cpu(var, cpu) (*RELOC_HIDE(&per_cpu__##var, __per_cpu_offset(cpu))) #define __get_cpu_var(var) (*RELOC_HIDE(&per_cpu__##var, __my_cpu_offset())) @@ -40,6 +45,8 @@ extern void setup_per_cpu_areas(void); #define DEFINE_PER_CPU(type, name) \ __typeof__(type) per_cpu__##name +#define DEFINE_PER_CPU_SHARED_CACHELINE_ALIGNED(type, name) \ + DEFINE_PER_CPU(type, name) #define per_cpu(var, cpu) (*((void)(cpu), &per_cpu__##var)) #define __get_cpu_var(var) per_cpu__##var diff -Nurp linux-2.6.21-rc7.0/include/asm-s390/percpu.h linux-2.6.21-rc7.1/include/asm-s390/percpu.h --- linux-2.6.21-rc7.0/include/asm-s390/percpu.h 2007-04-15 16:50:57.000000000 -0700 +++ linux-2.6.21-rc7.1/include/asm-s390/percpu.h 2007-05-04 11:20:19.000000000 -0700 @@ -41,6 +41,11 @@ extern unsigned long __per_cpu_offset[NR __attribute__((__section__(".data.percpu"))) \ __typeof__(type) per_cpu__##name +#define DEFINE_PER_CPU_SHARED_CACHELINE_ALIGNED(type, name) \ + __attribute__((__section__(".data.percpu.shared_cacheline_aligned"))) \ + __typeof__(type) per_cpu__##name \ + ____cacheline_aligned_in_smp + #define __get_cpu_var(var) __reloc_hide(var,S390_lowcore.percpu_offset) #define __raw_get_cpu_var(var) __reloc_hide(var,S390_lowcore.percpu_offset) #define per_cpu(var,cpu) __reloc_hide(var,__per_cpu_offset[cpu]) @@ -59,6 +64,8 @@ do { \ #define DEFINE_PER_CPU(type, name) \ __typeof__(type) per_cpu__##name +#define DEFINE_PER_CPU_SHARED_CACHELINE_ALIGNED(type, name) \ + DEFINE_PER_CPU(type, name) #define __get_cpu_var(var) __reloc_hide(var,0) #define __raw_get_cpu_var(var) __reloc_hide(var,0) diff -Nurp linux-2.6.21-rc7.0/include/asm-sparc64/percpu.h linux-2.6.21-rc7.1/include/asm-sparc64/percpu.h --- linux-2.6.21-rc7.0/include/asm-sparc64/percpu.h 2007-05-01 07:33:06.000000000 -0700 +++ linux-2.6.21-rc7.1/include/asm-sparc64/percpu.h 2007-05-04 11:20:51.000000000 -0700 @@ -17,6 +17,11 @@ extern unsigned long __per_cpu_shift; #define DEFINE_PER_CPU(type, name) \ __attribute__((__section__(".data.percpu"))) __typeof__(type) per_cpu__##name +#define DEFINE_PER_CPU_SHARED_CACHELINE_ALIGNED(type, name) \ + __attribute__((__section__(".data.percpu.shared_cacheline_aligned"))) \ + __typeof__(type) per_cpu__##name \ + ____cacheline_aligned_in_smp + register unsigned long __local_per_cpu_offset asm("g5"); /* var is in discarded region: offset to particular copy we want */ @@ -36,6 +41,8 @@ do { \ #define DEFINE_PER_CPU(type, name) \ __typeof__(type) per_cpu__##name +#define DEFINE_PER_CPU_SHARED_CACHELINE_ALIGNED(type, name) \ + DEFINE_PER_CPU(type, name) #define per_cpu(var, cpu) (*((void)cpu, &per_cpu__##var)) #define __get_cpu_var(var) per_cpu__##var diff -Nurp linux-2.6.21-rc7.0/include/asm-x86_64/percpu.h linux-2.6.21-rc7.1/include/asm-x86_64/percpu.h --- linux-2.6.21-rc7.0/include/asm-x86_64/percpu.h 2007-05-01 07:33:06.000000000 -0700 +++ linux-2.6.21-rc7.1/include/asm-x86_64/percpu.h 2007-05-04 11:21:25.000000000 -0700 @@ -20,6 +20,11 @@ #define DEFINE_PER_CPU(type, name) \ __attribute__((__section__(".data.percpu"))) __typeof__(type) per_cpu__##name +#define DEFINE_PER_CPU_SHARED_CACHELINE_ALIGNED(type, name) \ + __attribute__((__section__(".data.percpu.shared_cacheline_aligned"))) \ + __typeof__(type) per_cpu__##name \ + ____cacheline_internodealigned_in_smp + /* var is in discarded region: offset to particular copy we want */ #define per_cpu(var, cpu) (*({ \ extern int simple_identifier_##var(void); \ @@ -46,6 +51,8 @@ extern void setup_per_cpu_areas(void); #define DEFINE_PER_CPU(type, name) \ __typeof__(type) per_cpu__##name +#define DEFINE_PER_CPU_SHARED_CACHELINE_ALIGNED(type, name) \ + DEFINE_PER_CPU(type, name) #define per_cpu(var, cpu) (*((void)(cpu), &per_cpu__##var)) #define __get_cpu_var(var) per_cpu__##var