[Mesa-dev] [PATCH] intel: fix the wrong code to detect null texture.

[Kenneth Graunke]

On 09/08/2011 07:56 PM, Yuanhan Liu wrote:
> There is already comments show how to detect a null texture. Fix the
> code to match the comments.
> 
> This would fix the oglc divzero(basic.texQOrWEqualsZero) and
> divzero(basic.texTrivialPrim) test case fail.
> 
> Signed-off-by: Yuanhan Liu <yuanhan.liu at linux.intel.com>
> ---
>  src/mesa/drivers/dri/intel/intel_mipmap_tree.c |    2 +-
>  1 files changed, 1 insertions(+), 1 deletions(-)
> 
> diff --git a/src/mesa/drivers/dri/intel/intel_mipmap_tree.c b/src/mesa/drivers/dri/intel/intel_mipmap_tree.c
> index f36240d..717e0ae 100644
> --- a/src/mesa/drivers/dri/intel/intel_mipmap_tree.c
> +++ b/src/mesa/drivers/dri/intel/intel_mipmap_tree.c
> @@ -137,7 +137,7 @@ intel_miptree_create(struct intel_context *intel,
>     /*
>      * pitch == 0 || height == 0  indicates the null texture
>      */
> -   if (!mt || !mt->total_height) {
> +   if (!mt->total_width || !mt->total_height) {
>        free(mt);
>        return NULL;
>     }

I was a bit skeptical about removing the !mt check, but I think it's
actually okay since intel_miptree_create_internal never returns NULL.

I had to convince myself of that by writing the patch I'm about to reply
with.  intel_miptree_create_internal has explicit "if (!ok) return NULL"
code in it, but it turns out that ok is statically provable to always be
true.  My patch deletes it.

The only other reason it might want to return NULL is if allocation of
the struct intel_mipmap_tree fails, but it doesn't check for that.
Perhaps it should?  (But, it's a tiny struct and I'm pretty sure we fail
to check things like that all over the place, so...eh...)

Anyway, this looks right to me.  I'll defer to Eric and Ian though.

Reviewed-by: Kenneth Graunke <kenneth at whitecape.org>