On 27.01.2013 00:58, Eric Anholt wrote:
> Christoph Bumiller <e0425...@student.tuwien.ac.at> writes:
>> diff --git a/src/mesa/main/teximage.c b/src/mesa/main/teximage.c
>> index 31a559e..e71f6e1 100644
>> --- a/src/mesa/main/teximage.c
>> +++ b/src/mesa/main/teximage.c
>> +/** GL_ARB_texture_buffer_object */
>> +void GLAPIENTRY
>> +_mesa_TexBuffer(GLenum target, GLenum internalFormat, GLuint buffer)
>> +{
>> + struct gl_buffer_object *bufObj;
>> +
>> + GET_CURRENT_CONTEXT(ctx);
>> +
>> + if (!ctx->Extensions.ARB_texture_buffer_object) {
>> + _mesa_error(ctx, GL_INVALID_OPERATION, "glTexBuffer");
>> + return;
>> + }
> The check for also ctx->API == API_OPENGL_CORE here has been dropped, so
> I think the i965 driver would start accepting this function in compat
> contexts when it shouldn't.
>
> If that check gets re-added, this is
> Reviewed-by: Eric Anholt <e...@anholt.net>
To quote Ian on a reply to an earlier version of this, where I had the
check in the _mesa_TexBufferRange function also:
"
if (!(ctx->API == API_OPENGL_CORE &&
This check /shouldn't/ be necessary. Since Paul's rework of the
dispatch table code, the TexBufferRange function should only be put in
the dispatch table if the context is the correct profile. A trivial
piglit test will verify that: check that GL_INVALID_OPERATION is
generated if GL_ARB_texture_buffer_range is not supported and
glTexBufferRange(GL_TEXTURE_BUFFER, GL_RGBA, 0, 0, 0);
is called.
"
Why would this work for TexBufferRange and not TexBuffer non-Range ?
Thanks for reviewing,
Christoph
_______________________________________________
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/mesa-dev
