Commit Graph

460 Commits

Author SHA1 Message Date
01117c716d vkd3d-shader/hlsl: Remove unused expr ops.
These correspond to specific SM1 opcodes which we don't use (and which the native compiler never uses either).
2024-12-04 14:38:32 +01:00
74365417ac vkd3d-shader/hlsl: Output the candidates for ambiguous function calls. 2024-12-03 14:38:33 +01:00
198c3dc7b9 vkd3d-shader/hlsl: Remove hlsl_ir_vsir_instruction_ref, again. 2024-11-27 13:03:15 +01:00
1a6409cd5b vkd3d-shader/hlsl: Add parser support for stream-output object types.
Signed-off-by: Nikolay Sivov <nsivov@codeweavers.com>
2024-11-21 19:28:46 +01:00
cf27065b7d vkd3d-shader/hlsl: Add parser support for the RWByteAddressBuffer type.
Signed-off-by: Nikolay Sivov <nsivov@codeweavers.com>
2024-10-28 17:44:28 +01:00
9d82915629 vkd3d-shader/hlsl: Make allocation functions static again. 2024-10-24 20:50:59 +02:00
158bf794e6 vkd3d-shader/hlsl: Introduce hlsl_ir_vsir_instruction_ref, again.
This node type will be deleted (again) once the hlsl->vsir->tpf
translation is complete. It serves the purpose of allowing to keep
both real hlsl_ir_nodes and vsir_instructions in the hlsl_block,
until all the former can be translated into the latter.
2024-10-24 20:50:59 +02:00
cf3e251a9f vkd3d-shader/hlsl: Implement the f32tof16() intrinsic.
Signed-off-by: Nikolay Sivov <nsivov@codeweavers.com>
2024-10-22 20:28:33 +02:00
d562b03c43 vkd3d-shader/hlsl: Allow to force alignment on some semantic vars.
This is required to properly optimize signatures, because these
semantics must be alligned while being packed:
- Array elements.
- The first fields of structs.
- Major vectors of a matrix.

For now this has no effect since semantics are allocated with reg_size
4, but will have effect when optimizing interstage signatures.
2024-10-22 20:04:41 +02:00
7fd6c29ee8 vkd3d-shader/hlsl: Allow accounting for interpolation mode when allocating semantics.
For now this has no effect since semantics are allocated with reg_size
4, but will have effect when optimizing interstage signatures.
2024-10-22 20:03:39 +02:00
fb2b974466 vkd3d-shader/hlsl: Handle snorm/unorm types as resource formats.
Signed-off-by: Nikolay Sivov <nsivov@codeweavers.com>
2024-10-22 19:14:51 +02:00
df3aaac7cf vkd3d-shader/tpf: Introduce tpf_write_shader_function(). 2024-10-16 21:12:13 +02:00
7eee877dd4 vkd3d-shader/tpf: Make hlsl_sm4_register_from_semantic() independent of HLSL IR. 2024-10-16 21:07:51 +02:00
a243862b8c vkd3d-shader/tpf: Make sysval_semantic_from_hlsl() independent of HLSL IR. 2024-10-16 21:05:10 +02:00
7cc8151b81 vkd3d-shader/tpf: Replace uses of ctx->profile with tpf->program->shader_version. 2024-10-16 18:50:15 +02:00
5b21cc67f1 vkd3d-shader/hlsl: Split hlsl_sm4_write().
Similarly to the already done split from

    HLSL IR -> d3dbc

to

    HLSL IR -> vsir -> d3bc

we now start splitting the

    HLSL IR -> tpf

translation into

    HLSL IR -> vsir -> tpf

So hlsl_sm4_write is split into two functions, sm4_generate_vsir() and
tpf_compile().

This translation should be completed once tpf_compile() no longer needs
the hlsl_ctx and entry_func parameters.
2024-10-16 17:08:50 +02:00
3601397703 vkd3d-shader/d3dbc: Remove the "hlsl_" prefix from external functions.
They no longer rely on HLSL IR.

Also, rename them from "_from_semantic" to "from_semantic_name".
2024-10-16 16:52:45 +02:00
0c5dc53fd2 vkd3d-shader/hlsl: Allocate temporary registers separately for each entry function. 2024-10-15 16:39:50 +02:00
5f8570b933 vkd3d-shader/hlsl: Track whether a variable is read in any entry function. 2024-10-15 16:18:36 +02:00
f15a1c0b23 vkd3d-shader/hlsl: Record semantic extern vars separately for each entry function.
This is required in order to process the entry point function and the
patch constant function in hull shaders.
2024-10-15 16:18:27 +02:00
caa2a9d314 vkd3d-shader/hlsl: Parse sampler_state. 2024-10-02 22:07:34 +02:00
2722346ffb vkd3d-shader/hlsl: Do not lower index expressions for effects.
Signed-off-by: Nikolay Sivov <nsivov@codeweavers.com>
2024-10-02 22:05:24 +02:00
7e3231c749 vkd3d-shader/hlsl: Return an "error" expression when constructing an arithmetic expression from incompatible types. 2024-09-23 15:28:47 +02:00
c43e5c8eb5 vkd3d-shader/hlsl: Introduce the "error" type.
Currently, if an expression successfully parses according to the bison grammar,
but for one reason or another cannot generate a meaningful IR instruction, we
abort parsing with YYABORT. This includes, for example, an undefined variable or
function, invalid swizzle or field reference, or a constructor with a complex or
non-numeric data type.

Aborting parsing is unfortunate, however, because it means that any further
errors in the program cannot be caught by the programmer, increasing the number
of times they will need to fix errors and recompile.

The idea of this patch is that any such expression will instead generate an IR
node whose data type is of HLSL_CLASS_ERROR. Any further expression which would
consume an "error" typed instruction will instead immediately return an
expression of type "error" (probably the same one) instead of aborting or doing
any other type-checking.

Currently these "error" instructions should not pass the parsing stage, since
hlsl_compile_shader() will immediately notice that compilation has failed and
skip any optimization, lowering, or bytecode-writing.

A further direction to take this is to pre-allocate one "error" expression
immediately when creating the HLSL parser, and return that expression when we
fail to allocate an hlsl_ir_node of any type. This means we do not need to
handle allocation errors when constructing nodes, saving us quite a lot of error
handling (which is not only tedious but currently often broken, if nothing else
by virtue of neglecting cleanup of local variables).
2024-09-23 15:28:47 +02:00
8975933aff vkd3d-shader/hlsl: Store a pointer to the block's "value" instruction in the block. 2024-09-23 15:28:47 +02:00