Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > linux.kernel > #1241160 > unrolled thread

Re: [PATCH v2 2/3] word-at-a-time.h: support zero_bytemask() on alpha and tile

Back to article view | Back to linux.kernel

This discussion starts older than the indexed window; earlier articles aren't shown. The article labeled Started by below is the oldest one visible, not the original post.

Contents

  Re: [PATCH v2 2/3] word-at-a-time.h: support zero_bytemask() on  alpha and tile Ingo Molnar <mingo@kernel.org> - 2015-10-07 08:00 +0200
    Re: [PATCH v2 2/3] word-at-a-time.h: support zero_bytemask() on alpha  and tile Chris Metcalf <cmetcalf@ezchip.com> - 2015-10-07 16:00 +0200

#1241160 — Re: [PATCH v2 2/3] word-at-a-time.h: support zero_bytemask() on alpha and tile

* Chris Metcalf <cmetcalf@ezchip.com> wrote:

> Both alpha and tile needed implementations of zero_bytemask.
> 
> The alpha version is untested.
> 
> Signed-off-by: Chris Metcalf <cmetcalf@ezchip.com>
> ---
>  arch/alpha/include/asm/word-at-a-time.h | 2 ++
>  arch/tile/include/asm/word-at-a-time.h  | 8 +++++++-
>  2 files changed, 9 insertions(+), 1 deletion(-)
> 
> diff --git a/arch/alpha/include/asm/word-at-a-time.h b/arch/alpha/include/asm/word-at-a-time.h
> index 6b340d0f1521..902e6ab00a06 100644
> --- a/arch/alpha/include/asm/word-at-a-time.h
> +++ b/arch/alpha/include/asm/word-at-a-time.h
> @@ -52,4 +52,6 @@ static inline unsigned long find_zero(unsigned long bits)
>  #endif
>  }
>  
> +#define zero_bytemask(mask) ((2ul << (find_zero(mask) * 8)) - 1)

Small nit: please use a proper C inline function instead of CPP, as for example 
the PowerPC zero_bytemask() function is.

> -/* Generate 0x01 byte values for non-zero bytes using a SIMD instruction. */
> +/* Generate 0x01 byte values for zero bytes using a SIMD instruction. */
>  static inline unsigned long has_zero(unsigned long val, unsigned long *data,
>  				     const struct word_at_a_time *c)
>  {
> @@ -33,4 +33,10 @@ static inline long find_zero(unsigned long mask)
>  #endif
>  }
>  
> +#ifdef __BIG_ENDIAN
> +#define zero_bytemask(mask) (~1ul << (63 - __builtin_clzl(mask)))
> +#else
> +#define zero_bytemask(mask) ((2ul << __builtin_ctzl(mask)) - 1)
> +#endif

Ditto.

Other than that, for the whole series:

  Reviewed-by: Ingo Molnar <mingo@kernel.org>

Thanks,

	Ingo
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

[toc] | [next] | [standalone]

#1241520 — Re: [PATCH v2 2/3] word-at-a-time.h: support zero_bytemask() on alpha and tile

On 10/07/2015 01:53 AM, Ingo Molnar wrote:
> * Chris Metcalf <cmetcalf@ezchip.com> wrote:
>
>> Both alpha and tile needed implementations of zero_bytemask.
>>
>> The alpha version is untested.
>>
>> Signed-off-by: Chris Metcalf <cmetcalf@ezchip.com>
>> ---
>>   arch/alpha/include/asm/word-at-a-time.h | 2 ++
>>   arch/tile/include/asm/word-at-a-time.h  | 8 +++++++-
>>   2 files changed, 9 insertions(+), 1 deletion(-)
>>
>> diff --git a/arch/alpha/include/asm/word-at-a-time.h b/arch/alpha/include/asm/word-at-a-time.h
>> index 6b340d0f1521..902e6ab00a06 100644
>> --- a/arch/alpha/include/asm/word-at-a-time.h
>> +++ b/arch/alpha/include/asm/word-at-a-time.h
>> @@ -52,4 +52,6 @@ static inline unsigned long find_zero(unsigned long bits)
>>   #endif
>>   }
>>   
>> +#define zero_bytemask(mask) ((2ul << (find_zero(mask) * 8)) - 1)
> Small nit: please use a proper C inline function instead of CPP, as for example
> the PowerPC zero_bytemask() function is.

Indeed, I normally would do that.  But, I was trying to match the
existing style; there are nine definitions that are specified as macros
(including the one in asm-generic), and only one (powerpc 64-bit LE)
that was specified as an inline.

I did put together a v3 of the patch series that included all your
excellent suggestions, including the Reported-by, Tested-by, and
Reviewed-by tags, but then found out v2 was already pulled into
Linus's tree, so I guess that ship has sailed; oh well.

My inclination would be not to churn the tree with a patch to
convert those new zero_bytemask() macros to inlines, but I'm happy
to do so if you think it would be better.

-- 
Chris Metcalf, EZChip Semiconductor
http://www.ezchip.com

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

[toc] | [prev] | [standalone]

Back to top | Article view | linux.kernel

csiph-web