[PATCH 1/2] ARM: implement support for read-mostly sections

Catalin Marinas catalin.marinas at arm.com
Sun Dec 5 17:18:27 EST 2010


On 5 December 2010 11:43, Russell King - ARM Linux
<linux at arm.linux.org.uk> wrote:
> As our SMP implementation uses MESI protocols.  Grouping together data
> which is mostly only read together means that we avoid unnecessary
> cache line bouncing when this code shares a cache line with other data.
>
> In other words, cache lines associated with read-mostly data are
> expected to spend most of their time in shared state.
>
> Signed-off-by: Russell King <rmk+kernel at arm.linux.org.uk>
> ---
>  arch/arm/include/asm/cache.h  |    2 ++
>  arch/arm/kernel/vmlinux.lds.S |    1 +
>  2 files changed, 3 insertions(+), 0 deletions(-)
>
> diff --git a/arch/arm/include/asm/cache.h b/arch/arm/include/asm/cache.h
> index 9d61220..75fe66b 100644
> --- a/arch/arm/include/asm/cache.h
> +++ b/arch/arm/include/asm/cache.h
> @@ -23,4 +23,6 @@
>  #define ARCH_SLAB_MINALIGN 8
>  #endif
>
> +#define __read_mostly __attribute__((__section__(".data..read_mostly")))
> +
>  #endif
> diff --git a/arch/arm/kernel/vmlinux.lds.S b/arch/arm/kernel/vmlinux.lds.S
> index cead889..1581f6d 100644
> --- a/arch/arm/kernel/vmlinux.lds.S
> +++ b/arch/arm/kernel/vmlinux.lds.S
> @@ -167,6 +167,7 @@ SECTIONS
>
>                NOSAVE_DATA
>                CACHELINE_ALIGNED_DATA(32)
> +               READ_MOSTLY_DATA(32)

Should we change the alignments to 64?

-- 
Catalin



More information about the linux-arm-kernel mailing list