On Tue, Mar 25, 2003 at 12:17:44PM -0800, Gareth Hughes wrote:
> Umm, can you elaborate here? We (NVIDIA) don't support or expose > SGIS_fragment_specular_color. It's not even in the extension > registry...
[snip]
OpenGL extensions:
GL_EXT_abgr, GL_EXT_blend_color, GL_EXT_blend_logic_op, GL_EXT_blend_minmax, GL_EXT_blend_subtract, GL_EXT_convolution, GL_EXT_copy_texture, GL_EXT_histogram, GL_EXT_packed_pixels, GL_EXT_polygon_offset, GL_EXT_subtexture, GL_EXT_texture, GL_EXT_texture3D, GL_EXT_texture_object, GL_EXT_vertex_array, GL_SGI_color_matrix, GL_SGI_color_table, GL_SGI_texture_color_table, GL_SGIS_texture_filter4, GL_SGIS_detail_texture, GL_SGIS_texture_border_clamp, GL_SGIS_texture_select, GL_SGIS_texture_lod, GL_SGIX_list_priority, GL_SGIX_pixel_texture, GL_SGIX_texture_multi_buffer
[snip]
GL_SGI_color_table 0 glColorTableParameterfvSGI (nil) glColorTableParameterivSGI (nil) glColorTableSGI (nil) glCopyColorTableSGI (nil) glGetColorTableParameterfvSGI (nil) glGetColorTableParameterivSGI (nil) glGetColorTableSGI (nil)
which is not ok according to Brian's explanation.
Does this mean that glGetString(GL_EXTENSIONS) reported that GL_SGI_color_table was not supported? If that's the case then you either mis-read Brian's explaination or his mis-worded it slightly. If an extension is supported, then glXGetProcAddress must return a valid pointer. If the extension is not supported, the glXGetProcAddress may or may not return NULL.
If you think about it, that makes sense. You have a libGL.so from a week ago. Today somebody releases an new extension called GL_FOO_funky_chicken. How could the libGL.so on your system possibly know anything about the functions in that extension?
On the other hand, if glGetString(GL_EXTENSIONS) reports that the extension is supported, but glXGetProcAddress returns NULL, then there is a bug in libGL.so. Basically, the server has a list of GL extensions that it supports, and the libGL.so has a list that it supports. The string returned by glGetString is supposed to be the intersection of the two. I don't think this is a path that gets a lot of testing
Given that GL_SGI_color_table is listed in the glxinfo output you show, I'm guessing it's the later. In that case, it sounds like you've found a bug in Nvidia's libGL. :)
------------------------------------------------------- This SF.net email is sponsored by: The Definitive IT and Networking Event. Be There! NetWorld+Interop Las Vegas 2003 -- Register today! http://ads.sourceforge.net/cgi-bin/redirect.pl?keyn0001en _______________________________________________ Dri-devel mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/dri-devel
