vkd3d-shader: Maintenance patches
Merge request reports
Activity
added 43 commits
-
84a6cea5...e2dac061 - 38 commits from branch
wine:master
- 4a8c044e - vkd3d-shader/d3dbc: Return an error from vkd3d_shader_sm1_parser_create() if the parser failed.
- 737061ec - vkd3d-shader/tpf: Return an error from vkd3d_shader_sm4_parser_create() if the parser failed.
- 1b01b6d3 - vkd3d-shader: Introduce an internal sm6 signature structure.
- 10091e5e - tests/d3d12: Test register relative addressing in vertex and pixel shaders.
- 50fedad4 - vkd3d-shader/tpf: Fail parsing if an input/output parameter order is > 2.
Toggle commit list-
84a6cea5...e2dac061 - 38 commits from branch
added 4 commits
- 8e42b7c3 - vkd3d-shader/d3dbc: Return an error from vkd3d_shader_sm1_parser_create() if the parser failed.
- 7d4c2008 - vkd3d-shader/tpf: Return an error from vkd3d_shader_sm4_parser_create() if the parser failed.
- 35c119cd - vkd3d-shader: Introduce an internal sm6 signature structure.
- 8b25e7c5 - tests/d3d12: Test register relative addressing in vertex and pixel shaders.
Toggle commit listmentioned in merge request !198 (merged)
775 775 VKD3D_SIV_LINE_DENSITY_TESS_FACTOR = 22, 776 776 }; 777 777 778 struct sm6_signature_element 779 { 780 const char *semantic_name; 781 unsigned int semantic_index; 782 unsigned int stream_index; 783 enum vkd3d_shader_sysval_semantic sysval_semantic; 784 enum vkd3d_shader_component_type component_type; 785 unsigned int register_index; 786 unsigned int register_count; 787 unsigned int mask; 788 unsigned int used_mask; 789 enum vkd3d_shader_minimum_precision min_precision; 790 }; - Comment on lines +778 to +790
Is it really sensible to call this
sm6_signature_element
? AFAIU it is not specific for SM6, it is used for all shader models. Maybeinternal_signature_element
would be a more neutral name?Relatedly, why isn't the new field exposed to the public API? Just because we don't want to go through the hassle of updating the public API, or is it genuinely uninteresting for our clients?
Is it really sensible to call this
sm6_signature_element
? AFAIU it is not specific for SM6, it is used for all shader models. Maybeinternal_signature_element
would be a more neutral name?True. We wouldn't need the "internal_" prefix either though, because the public version is called "vkd3d_shader_signature_element".
Relatedly, why isn't the new field exposed to the public API? Just because we don't want to go through the hassle of updating the public API, or is it genuinely uninteresting for our clients?
Mostly some variant of the former; that doesn't prevent us from exposing this at some point in the future, of course. I think we'd want to wait with exposing this in the public API until broader DXIL support is settled a bit, in case there are additional API changes we'd like to make.
Is there a reason why this couldn't be a
.shader_test
, beside missing HLSL compiler support?I don't think so, no; see also !181 (comment 31337)
added 44 commits
-
9bdb55ca...34ddc133 - 40 commits from branch
wine:master
- 85eb2314 - vkd3d-shader/d3dbc: Return an error from vkd3d_shader_sm1_parser_create() if the parser failed.
- 5ae06816 - vkd3d-shader/tpf: Return an error from vkd3d_shader_sm4_parser_create() if the parser failed.
- a0a18b16 - vkd3d-shader: Introduce an internal shader signature structure.
- 7917a682 - tests/d3d12: Test register relative addressing in vertex and pixel shaders.
Toggle commit list-
9bdb55ca...34ddc133 - 40 commits from branch