On Mon, Nov 21, 2011 at 8:01 PM, Ian Romanick <i...@freedesktop.org> wrote: > On 11/19/2011 07:42 AM, Marek Olšák wrote: >> >> On Fri, Nov 18, 2011 at 9:44 PM, Ian Romanick<i...@freedesktop.org> wrote: >>> >>> On 11/18/2011 11:27 AM, Marek Olšák wrote: >>> >>> This patch also needs to change the _mesa_glsl_supported_extensions table >>> in >>> glsl_parser_extras.cpp. AMD_conservative_depth is used for both versions >>> of >>> the extension in the table. >> >> But that refers to the single extension flag in mtypes.h and there's >> only one: the AMD one. Not sure what else I should change in that >> table. > > You are correct. Looking closer at the EXT macro used to generate that > table, I see why there are two enable bits in _mesa_glsl_parse_state. The > structure field name and the string come from the same macro parameter. > >>> I'm not super convinced that we even need separate enable flags. Both >>> extensions add the exact same functionality using the exact same layout >>> qualifiers. It's not a big deal to me either way, though. >> >> I guess this would end up being incorrect with unified flags: >> >> #extension GL_AMD_conservative_depth : enable >> #extension GL_ARB_conservative_depth : disable // would disable AMD too > > That's an interesting case. What does AMD's closed-source driver do in that > case? I think that's the only other shipping implementation of the AMD > extensions, so as long as we do what they do...
Sorry I don't have the AMD driver right now. Marek _______________________________________________ mesa-dev mailing list mesa-dev@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/mesa-dev