[Mesa-dev] [PATCH 3/8] mesa: Add interpolation override bitfields.

[Brian Paul]

On 10/24/2011 05:47 PM, Paul Berry wrote:
> On 24 October 2011 16:37, Paul Berry <stereotype441 at gmail.com
> <mailto:stereotype441 at gmail.com>> wrote:
>
>     On 24 October 2011 15:58, Brian Paul <brianp at vmware.com
>     <mailto:brianp at vmware.com>> wrote:
>
>         On 10/24/2011 03:38 PM, Paul Berry wrote:
>
>             This patch adds the bitfields InterpOverridesFlat,
>             InterpOverridesSmooth, and InterpOverridesNoperspective to
>             gl_fragment_program.  These bitfields keep track of which
>             fragment
>             shader inputs are overridden with the GLSL "flat",
>             "smooth", and
>             "noperspective" interpolation qualifiers.
>
>
>         The names of those fields seems a little confusing to me.  For
>         example, "InterpOverridesFlat" sounds like a field that
>         overrides flat shading with something else.
>
>         How about just "InterpFlat", "InterpSmooth", etc?
>
>
>     The meaning I was trying to convey with "overrides" is that bits
>     are only set in these bitfields if the interpolation mode is
>     overridden in GLSL.  For fragment shader inputs that don't have
>     their interpolation mode overridden, all of the bitfields have a
>     zero.  (I had to do this in order to avoid adding a bunch of code
>     to the handling of fixed function fragment shading and ARB
>     fragment programs).
>
>
> Another thought: how would you feel about "InterpQualifiedFlat",
> "InterpQualifiedSmooth", etc?  That would satisfy my desire to clarify
> that these bits are only set when interpolation is overridden in GLSL,
> but without the potential for the misinterpretation you were concerned
> about.

If we stick with bitfields, that looks better to me.

-Brian