[Mesa-dev] [PATCH] include: update GL & GLES headers

[Ian Romanick]

On 08/02/2018 04:40 PM, Ilia Mirkin wrote:
> On Thu, Aug 2, 2018 at 7:34 PM, Ian Romanick <idr at freedesktop.org> wrote:
>> On 08/02/2018 02:26 PM, Sagar Ghuge wrote:
>>> I have few comments below, I might be wrong about those but except that
>>> this patch looks good to me. I checked it against Khronos headers.
>>>
>>> On 08/01/2018 11:34 AM, Marek Olšák wrote:
>>>>   +#ifndef GL_AMD_framebuffer_multisample_advanced
>>>> +#define GL_AMD_framebuffer_multisample_advanced
>>>
>>> I might be wrong but value is missing for a macro I guess. I am not sure
>>> but, are we just trying to expand it to nothing. Please let me know if I
>>> am missing something here.
>>
>> This is the same as '#define GL_AMD_framebuffer_multisample_advanced 1'.
> 
> For the sake of completeness, it's actually different.
> 
> If you #define foo 1, then the token "foo" will be replaced with the
> token "1" after preprocessing. If you just do "#define foo", then foo
> may only be used with #ifdef foo, or #if defined(foo). Using foo
> outside of that will cause it to be replaced with nothing.

I was going to argue, but...

[idr at localhost home]$ echo "#define foo
foo" | cpp -E
# 1 "<stdin>"
# 1 "<built-in>"
# 1 "<command-line>"
# 31 "<command-line>"
# 1 "/usr/include/stdc-predef.h" 1 3 4
# 32 "<command-line>" 2
# 1 "<stdin>"


[idr at localhost home]$

I think this is a (really trivial) bug in the Khronos header.  Boo.

> [Your other point about auto-generation and coming from Khronos
> obviously stand.]
> 
>   -ilia