[Mesa-dev] [PATCH 2/2] mesa: utilize context version routines

[Brian Paul]

On 09/03/2012 11:35 AM, Eric Anholt wrote:
> Kenneth Graunke<kenneth at whitecape.org>  writes:
>
>> I'm really ambivalent about these patches.
>>
>> 1. I'm not a huge fan of the name "have_version"...it sounds like it
>> would return whether a driver supports a given version, not whether the
>> current context's version is a certain value.
>>
>> 2. Personally I think ctx->Version<= XY is clearer than
>> !_mesa_have_version(ctx, X, Y + 1).  With the two-digit notation, you
>> can just do whatever comparison you like, rather than having to negate
>>> = and possibly increment the minor version being compared.
>>
>> 3. _mesa_have_version(ctx, 3, 1) is longer than ctx->Version>= 31
>>
>> 3. I really don't see the major * 10 + minor notation needing to be
>> changed in the future.  Even if Khronos did offer (say) a GL 4.3.1
>> release, the likelihood of it making incompatible changes over 4.3 that
>> require special checks is infinitesimal.  It would just be clarifications...
>>
>> Normally, I'm all for encapsulation, but I guess I just don't see much
>> point.  That said, I won't object too strongly if people prefer this
>> approach.
>
> Seconding this.

I don't have a strong opinion on this bike shed either.

Maybe just leave things as-is for now.  This patch can always be 
reused in the future if the situation changes.

-Brian