hi:
2011/2/18 Anoop P A <anoop.pa@xxxxxxxxx>:
> On Fri, 2011-02-18 at 15:12 +0800, loody wrote:
>> Hi all:
>>
>> 2011/2/18 John Mahoney <jmahoney@xxxxxxxx>:
>> > On Thu, Feb 17, 2011 at 9:17 AM, loody <miloody@xxxxxxxxx> wrote:
>> >> hi :-)
>> >>
>> >> 2011/2/16 Mulyadi Santosa <mulyadi.santosa@xxxxxxxxx>:
>> >>> Hi :)
>> >>>
>> >>> On Wed, Feb 16, 2011 at 12:59, Rajat Jain <rajatjain@xxxxxxxxxxx> wrote:
>> >>>> Hello loody,
>> >>>>
>> >>>>> 1. in kernel/trace, I always see "__read_mostly" at the end of
>> >>>>> parameter is that a compiler optimization parameter?
>> >>>>
>> >>>> Yes, it is a hint to the compiler that the parameter is mostly read, thus if the compiler has to make a decision between optimizing one of the read / write paths, it will optimize the read path even at the expense of write path.
>> >>>
>> >>>
>> >>> To be precise, they will be grouped into same cache line as much as
>> >>> possible. By doing so, those cache line won't be invalidated so often
>> >>> (keeping them "hot" :) hehehhe )
>> >>
>> >> I cannot find it on the gcc manual.
>> >> is it a option in kernel for kernel usage?
>> >> if so, where I can found them.
>> >> If not, can I use it on normal user level program?
>> >>
>> >
>> > It is a macro defined for x86 as:
>> >
>> > #define __read_mostly __attribute__((__section__(".data..read_mostly")))
>> I try to find "__read_mostly" of mips arch and below are the results I
>> grep on arch/
>>
>> ......
>> ia64/xen/hypervisor.c:29:struct shared_info *HYPERVISOR_shared_info
>> __read_mostly =
>> mips/kernel/irq_txx9.c:58:static struct txx9_irc_reg __iomem
>> *txx9_ircptr __read_mostly;
>> mips/kernel/irq_txx9.c:63:} txx9irq[TXx9_MAX_IR] __read_mostly;
>> mips/kernel/setup.c:35:struct cpuinfo_mips cpu_data[NR_CPUS] __read_mostly;
>> mips/kernel/setup.c:55:unsigned long mips_machtype __read_mostly = MACH_UNKNOWN;
>> mips/kernel/setup.c:72:const unsigned long mips_io_port_base __read_mostly = -1;
>> mips/kernel/smp.c:58:cpumask_t cpu_sibling_map[NR_CPUS] __read_mostly;
>> mips/kernel/process.c:329:static struct mips_frame_info schedule_mfi
>> __read_mostly;
>> mips/txx9/generic/irq_tx4939.c:50:} tx4939irq[TX4939_NUM_IR] __read_mostly;
>> mips/mm/c-r4k.c:67:static unsigned long icache_size __read_mostly;
>> mips/mm/c-r4k.c:68:static unsigned long dcache_size __read_mostly;
>> mips/mm/c-r4k.c:69:static unsigned long scache_size __read_mostly;
>> parisc/mm/init.c:543:unsigned long *empty_zero_page __read_mostly;
>> parisc/include/asm/cache.h:31:#define __read_mostly
>> __attribute__((__section__(".data.read_mostly")))
>> .....
>> x86/vdso/vdso32-setup.c:57:unsigned int __read_mostly vdso_enabled =
>> VDSO_DEFAULT;
>> x86/include/asm/cache.h:10:#define __read_mostly
>> __attribute__((__section__(".data.read_mostly")))
>>
>>
>> Does Mips arch not use this option?
>
> arch/mips/include/asm/cache.h:#define __read_mostly
> __attribute__((__section__(".data.read_mostly")))
>
>
>
That is wired.
I cat the content of my cache.h as below:
linux-2.6.33.4.mips# cat arch/mips/include/asm/cache.h
/*
* This file is subject to the terms and conditions of the GNU General Public
* License. See the file "COPYING" in the main directory of this archive
* for more details.
*
* Copyright (C) 1997, 98, 99, 2000, 2003 Ralf Baechle
* Copyright (C) 1999 Silicon Graphics, Inc.
*/
#ifndef _ASM_CACHE_H
#define _ASM_CACHE_H
#include <kmalloc.h>
#define L1_CACHE_SHIFT CONFIG_MIPS_L1_CACHE_SHIFT
#define L1_CACHE_BYTES (1 << L1_CACHE_SHIFT)
#define SMP_CACHE_SHIFT L1_CACHE_SHIFT
#define SMP_CACHE_BYTES L1_CACHE_BYTES
#endif /* _ASM_CACHE_H */
linux-2.6.33.4.mips#
Can I get the conclusion that before 2.6.33.4, there is no "__read_mostly"?
Thanks a lot,
miloody
