[Mesa-dev] [PATCH 01/20] mesa/format_utils: Fix a bug in one of the format helper functions

[Michel Dänzer]

On 18.11.2014 17:43, Iago Toral Quiroga wrote:
> From: Jason Ekstrand <jason.ekstrand at intel.com>

The commit short log (first line of the commit log) should at least say 'snorm_to_float' instead of 'one of the format helper functions'. It could probably have a more informative summary of what the change does as well.


> ---
>   src/mesa/main/format_utils.c | 2 +-
>   1 file changed, 1 insertion(+), 1 deletion(-)
> 
> diff --git a/src/mesa/main/format_utils.c b/src/mesa/main/format_utils.c
> index 93a0cea..b6d0fbc 100644
> --- a/src/mesa/main/format_utils.c
> +++ b/src/mesa/main/format_utils.c
> @@ -152,7 +152,7 @@ unorm_to_float(unsigned x, unsigned src_bits)
>   static inline float
>   snorm_to_float(int x, unsigned src_bits)
>   {
> -   if (x == -MAX_INT(src_bits))
> +   if (x < -MAX_INT(src_bits))
>         return -1.0f;
>      else
>         return x * (1.0f / (float)MAX_INT(src_bits));

Should it be if (x <= -MAX_INT(src_bits))? Even if the else case
is guaranteed to produce -1.0f for x == -MAX_INT(src_bits), it seems like wasted effort.


-- 
Earthling Michel Dänzer               |               http://www.amd.com
Libre software enthusiast             |             Mesa and X developer