[Mesa-dev] [PATCH 1/2][RFC] docs: Update with GLES3.2 entries and status

[Thomas Helland]

2015-08-12 7:45 GMT+02:00 Tapani Pälli <tapani.palli at intel.com>:
>
>
> On 08/11/2015 08:48 PM, Thomas Helland wrote:
>>
>> Signed-off-by: Thomas Helland <thomashelland90 at gmail.com>
>> ---
>> These are listed to the best of my knowledge, looking
>> at the nvidia driver dropped yesterday, and a glance at the
>> "what changed according to the ARB variant"
>> section of each extension spec. Hopefully it should not
>> be too far off. Feel free to leave your comments
>>
>>   docs/GL3.txt | 25 +++++++++++++++++++++++++
>>   1 file changed, 25 insertions(+)
>>
>> diff --git a/docs/GL3.txt b/docs/GL3.txt
>> index 8124383..2148ca0 100644
>> --- a/docs/GL3.txt
>> +++ b/docs/GL3.txt
>> @@ -223,6 +223,31 @@ GLES3.1, GLSL ES 3.1
>>     GS5 Packing/bitfield/conversion functions            DONE (i965, nvc0,
>> r600, radeonsi)
>>     GL_EXT_shader_integer_mix                            DONE (all drivers
>> that support GLSL)
>>
>> +GLES3.2, GLSL ES 3.2
>> +  GL_EXT_color_buffer_float                            DONE (all drivers)
>> +  GL_KHR_blend_equation_advanced                       not started
>> +  GL_KHR_debug                                         DONE (all drivers)
>> +  GL_KHR_robustness                                    90% done (the ARB
>> variant)
>> +  GL_KHR_texture_compression_astc_hdr                  in progress
>> (Nanley Chery)
>> +  GL_OES_copy_image                                    60% done (based on
>> parts of GL_ARB_copy_image, which is done for some drivers)
>> +  GL_OES_draw_buffers_indexed                          not started
>> +  GL_OES_draw_elements_base_vertex                     90% done (based on
>> GL_ARB_draw_elements_base_verte, which is done for all drivers)
>> +  GL_OES_geometry_shader                               80% done (based on
>> GL_ARB_geometry_shader4, which is done for all drivers)
>> +  GL_OES_gpu_shader5                                   90% done (based on
>> parts of GL_ARB_gpu_shader5, which is done for some drivers)
>> +  GL_OES_primitive_bounding box                        not started
>> +  GL_OES_sample_shading                                90% done (based on
>> parts of GL_ARB_sample_shading, which is done for some drivers)
>> +  GL_OES_sample_variables                              90% done (based on
>> parts of GL_ARB_sample_shading, which is done for some drivers)
>> +  GL_OES_shader_image_atomic                           90% done (based on
>> parts of GL_ARB_shader_image_load_store, which is done for some drivers)
>> +  GL_OES_shader_io_blocks                              not started
>> +  GL_OES_shader_multisample_interpolation              started (based on
>> parts of GL_ARB_gpu_shader5, which is done)
>> +  GL_OES_tessellation_shader                           90% done (based on
>> GL_ARB_tessellation_shader, which is done for some drivers)
>> +  GL_OES_texture_border_clamp                          90% done (the ARB
>> variant is done)
>> +  GL_OES_texture_buffer                                90% done
>> (combination of ARB_texture_buffer_object, ARB_texture_buffer_range, and
>> ARB_texture_buffer_object_rgb32 that are all done)
>> +  GL_OES_texture_cube_map_array                        90% done (based on
>> GL_ARB_texture_cube_map_array, which is done for all drivers)
>> +  GL_OES_texture_stencil8                              90% done (based on
>> parts of GL_ARB_texture_stencil8, which is done for some drivers)
>> +  GL_OES_texture_storage_multisample_2d_array          not started
>
>
>
> BTW I've started GL_OES_texture_storage_multisample_2d_array, it's available
> at
>
> http://cgit.freedesktop.org/~tpalli/mesa/log/?h=OES_texture_storage_ms_2d_array
>

Ok, I'll add OES_texture ......... in progress (Tapani Pälli)

>
> In general, I would prefer 'not started' instead of a percentage for these
> new entries. These values give impression that someone is actually working
> on item and is going to finish it soon?
>

I was debating with myself what to do. My thought was that it sheds some
light on how "finished" it is since we've got the ARB variant. But since I've
decided to also list what ARB variants the OES is based on, and the ARB
variants status, I guess there is no need to list a percentage.
I'll drop the percentage, and just say "not started" with the comment
about the ARB-variants status as it stands now.

> However if others don't mind then I guess it's ok. Extension list looks
> correct for me, for the series:
>
> Reviewed-by: Tapani Pälli <tapani.palli at intel.com>
>

Thanks. I'll let it soak for a couple days before getting it pushed.

>
>> +
>> +
>>     Additional functions not covered above:
>>         glMemoryBarrierByRegion
>>         glGetTexLevelParameter[fi]v - needs updates to restrict to GLES
>> enums
>>
>