[PATCH 1/2] ARM: dcache: select DCACHE_WORD_ACCESS for big-endian CPUs

[Nicolas Pitre]

On Tue, 17 Dec 2013, Will Deacon wrote:

> With commit 11ec50caedb5 ("word-at-a-time: provide generic big-endian
> zero_bytemask implementation"), the asm-generic word-at-a-time code now
> provides a zero_bytemask implementation, allowing us to make use of
> DCACHE_WORD_ACCESS on big-endian CPUs, providing our
> load_unaligned_zeropad function is endianness-clean.
> 
> This patch reworks the load_unaligned_zeropad fixup code to work for
> both big- and little-endian CPUs, then removes the !CPU_BIG_ENDIAN check
> when selecting DCACHE_WORD_ACCESS.
> 
> Signed-off-by: Will Deacon <will.deacon at arm.com>

Small comment below.

> ---
>  arch/arm/Kconfig                      |  2 +-
>  arch/arm/include/asm/word-at-a-time.h | 18 ++++++++++--------
>  2 files changed, 11 insertions(+), 9 deletions(-)
> 
> diff --git a/arch/arm/Kconfig b/arch/arm/Kconfig
> index 88f726d36a40..38ccfd0adbf9 100644
> --- a/arch/arm/Kconfig
> +++ b/arch/arm/Kconfig
> @@ -11,7 +11,7 @@ config ARM
>  	select BUILDTIME_EXTABLE_SORT if MMU
>  	select CLONE_BACKWARDS
>  	select CPU_PM if (SUSPEND || CPU_IDLE)
> -	select DCACHE_WORD_ACCESS if (CPU_V6 || CPU_V6K || CPU_V7) && !CPU_BIG_ENDIAN && MMU
> +	select DCACHE_WORD_ACCESS if (CPU_V6 || CPU_V6K || CPU_V7) && MMU
>  	select GENERIC_ATOMIC64 if (CPU_V7M || CPU_V6 || !CPU_32v6K || !AEABI)
>  	select GENERIC_CLOCKEVENTS_BROADCAST if SMP
>  	select GENERIC_IDLE_POLL_SETUP
> diff --git a/arch/arm/include/asm/word-at-a-time.h b/arch/arm/include/asm/word-at-a-time.h
> index 4d52f92967a6..a6d0a29861e7 100644
> --- a/arch/arm/include/asm/word-at-a-time.h
> +++ b/arch/arm/include/asm/word-at-a-time.h
> @@ -48,10 +48,14 @@ static inline unsigned long find_zero(unsigned long mask)
>  	return ret;
>  }
>  
> -#ifdef CONFIG_DCACHE_WORD_ACCESS
> -
>  #define zero_bytemask(mask) (mask)
>  
> +#else	/* __ARMEB__ */
> +#include <asm-generic/word-at-a-time.h>
> +#endif
> +
> +#ifdef CONFIG_DCACHE_WORD_ACCESS
> +
>  /*
>   * Load an unaligned word from kernel space.
>   *
> @@ -73,7 +77,11 @@ static inline unsigned long load_unaligned_zeropad(const void *addr)
>  	"	bic	%2, %2, #0x3\n"
>  	"	ldr	%0, [%2]\n"
>  	"	lsl	%1, %1, #0x3\n"
> +#ifndef __ARMEB__
>  	"	lsr	%0, %0, %1\n"
> +#else
> +	"	lsl	%0, %0, %1\n"
> +#endif

You could get rid of the #ifdef by using:

		pull	%0, %0, %1

Except this is not a pure assembly file and can't rely on 
<asm/assembler.h>.  Bah.  In that case...

Acked-by: Nicolas Pitre <nico at linaro.org>


Nicolas