Commit Graph

5412 Commits

Author SHA1 Message Date
Giovanni Mascellani
3dc7f322cc vkd3d-shader/ir: Validate INCONTROLPOINT destination parameters. 2024-12-17 16:39:32 +01:00
Giovanni Mascellani
7d87b4e869 vkd3d-shader/ir: Validate OUTPUT destination parameters. 2024-12-17 16:39:32 +01:00
Giovanni Mascellani
b194e5dc4e vkd3d-shader/ir: Validate PATCHCONST destination parameters using a uniform helper. 2024-12-17 16:39:32 +01:00
Giovanni Mascellani
38a5c905db vkd3d-shader/ir: Validate INPUT destination parameters using a uniform helper. 2024-12-17 16:39:32 +01:00
Giovanni Mascellani
8d4b790eb2 vkd3d-shader/ir: Represent information about I/O registers with a table.
It makes the code quite longer, but also easier to read and extend
with further properties.

A (desirable) side effect of this commit is that it is checked
whether I/O register types are legal depending on the shader type
and phase, while before that was assumed.

This commit introduces enum vsir_io_reg_type and enum vsir_phase
which shadow enum vkd3d_shader_register_type and enum
vkd3d_shader_opcode, with the goal of making the data tables
smaller.
2024-12-17 16:39:32 +01:00
Elizabeth Figura
de615609dc vkd3d-shader/d3dbc: Use struct vkd3d_shader_instruction instead of struct sm1_instruction. 2024-12-16 17:11:57 +01:00
Elizabeth Figura
77ac13e6b9 vkd3d-shader/d3dbc: Use struct vkd3d_shader_src_param instead of struct sm1_src_register. 2024-12-16 17:11:57 +01:00
Elizabeth Figura
1e43088662 vkd3d-shader/d3dbc: Use struct vkd3d_shader_dst_param instead of struct sm1_dst_register. 2024-12-16 17:11:57 +01:00
Henri Verbeet
c4f69f4d3d vkd3d-shader/hlsl: Prefer overload candidates without component count widening. 2024-12-16 17:09:40 +01:00
Henri Verbeet
64bb4150f0 vkd3d-shader/hlsl: Prefer overload candidates without component type narrowing. 2024-12-16 17:09:40 +01:00
Henri Verbeet
32d432ab5e vkd3d-shader/hlsl: Prefer overload candidates with matching component types. 2024-12-16 17:09:40 +01:00
Henri Verbeet
6b8878377f vkd3d-shader/hlsl: Prefer overload candidates with matching component type classes. 2024-12-16 17:09:40 +01:00
Henri Verbeet
63fce3062e vkd3d-shader/hlsl: Prefer overload candidates without component count narrowing. 2024-12-16 17:09:40 +01:00
Giovanni Mascellani
dd0ed989a1 vkd3d-shader/d3dbc: Normalise I/O register write masks when not disassembling.
Sometimes SM1-3 shaders contain write masks that exceed the
signature element masks. That happens because SM1-3 shaders do not
have a concept of signature and signature masks, and OTOH aren't
always able to express any given write mask.

In VSIR we don't want to deal with I/O register masks exceeding the
corresponding signature element mask or usage mask, because, for
instance, for higher shader models it can complicate dealing with
DCL_INDEX_RANGE. In order to have uniform rules for all shader
models we normalise masks coming from SM1-3 shaders.

We don't do that normalisation when disassembling, in order to
preserve the expected output.
2024-12-12 17:21:49 +01:00
Giovanni Mascellani
64126a00c3 vkd3d-shader/ir: Give more meaningful names to I/O normalisation levels.
The previous names "not normalised" and "fully normalised" have meanings
which are likely to change with time. OTOH including a description of the
normalisation level in the enumerant seems excessive. Relating
normalisation levels to shader model versions might be a reasonable
compromise.
2024-12-12 17:21:49 +01:00
Elizabeth Figura
09095403ac vkd3d-shader/hlsl: Store swizzles in vsir format.
The previous format matched sm1 and sm4, but if we're going to be feeding
everything through vsir, we want vsir's format.
2024-12-12 17:20:23 +01:00
Elizabeth Figura
474a0ac4e2 vkd3d-shader/hlsl: Use swizzle helpers in more places. 2024-12-12 17:11:53 +01:00
Elizabeth Figura
a905a78a96 vkd3d-shader/hlsl: Use a structure for matrix swizzles. 2024-12-12 17:11:27 +01:00
Elizabeth Figura
70e2148fcb vkd3d-shader/hlsl: Document matrix swizzles. 2024-12-12 16:56:52 +01:00
Victor Chiletto
ed6061dfae vkd3d-shader/hlsl: Drop the _ir_ infix from enum hlsl_loop_unroll_type. 2024-12-12 16:54:07 +01:00
Victor Chiletto
a1d995e740 vkd3d-shader/hlsl: Unroll loops with conditional jumps. 2024-12-12 16:54:04 +01:00
Victor Chiletto
351d58a95b vkd3d-shader/hlsl: Partially defer continue resolution.
We need 'for' iter blocks to be separate for loop unrolling.
2024-12-12 16:41:34 +01:00
Victor Chiletto
5d8448a44e vkd3d-shader/hlsl: Remove loop_unrolling_find_unrollable_loop().
Since loop unrolling doesn't clone the entire program anymore, we can
express it as a hlsl_transform_ir() transform.
2024-12-12 16:41:34 +01:00
Victor Chiletto
de3a365fea vkd3d-shader/hlsl: Do not clone the entire program for loop unrolling. 2024-12-12 16:41:34 +01:00
Victor Chiletto
0a8c4a6fa2 vkd3d-shader/hlsl: Run copy propagation incrementally during loop unrolling. 2024-12-12 16:41:34 +01:00