18 Commits

Author SHA1 Message Date
Nikita Popov
1128a4fd2c
[HLSL] Don't use CreateRuntimeFunction for intrinsics (#145334)
HLSL uses CreateRuntimeFunction for three intrinsics. This is pretty
unusual thing to do, and doesn't match what the rest of the file does.

I suspect this might be because these are convergent calls, but the
intrinsics themselves are already marked convergent, so it's not
necessary for clang to manually add the attribute.

This does lose the spir_func CC on the intrinsic declaration, but again,
CC should not be relevant to intrinsics at all.
2025-06-23 17:37:33 +02:00
Steven Perron
01d648a429
[HLSL][SPIRV] Reapply "[HLSL][SPIRV] Add vk::constant_id attribute." (#144902)
- **Reapply "[HLSL][SPIRV] Add vk::constant_id attribute." (#144812)**
- **Fix memory leak.**
2025-06-19 11:52:55 -04:00
Steven Perron
5f69d680e2
Revert "[HLSL][SPIRV] Add vk::constant_id attribute." (#144812)
Reverts llvm/llvm-project#143544
2025-06-18 19:30:43 -04:00
Steven Perron
acde20b560
[HLSL][SPIRV] Add vk::constant_id attribute. (#143544)
The vk::constant_id attribute is used to indicate that a global const
variable
represents a specialization constant in SPIR-V. This PR adds this
attribute to clang.

The documentation for the attribute is
[here](https://github.com/microsoft/DirectXShaderCompiler/blob/main/docs/SPIR-V.rst#specialization-constants).

The strategy is to to modify the initializer to get the value of a
specialize constant for a builtin defined in the SPIR-V backend.

Implements https://github.com/llvm/wg-hlsl/pull/287

Fixes https://github.com/llvm/llvm-project/issues/142448

---------

Co-authored-by: Nathan Gauër <github@keenuts.net>
2025-06-18 06:39:52 -04:00
Steven Perron
bd33eef7f1
[HLSL][SPIRV] Use resource names (#143412)
The SPIR-V backend does not have access to the original name of a
resource in the source, so it tries to create a name. This leads to some
problems with reflection.
    
That is why start to pass the name of the resource from Clang to the
SPIR-V backend.
    
Fixes #138533
2025-06-13 12:21:38 -04:00
Nathan Gauër
6582d7d348
[HLSL] Add WaveGetLaneCount() intrinsic to FE (#143127)
This commit adds code to lower WaveGetLaneCount() into the SPV or DXIL
intrinsic. The backends will then need to lower the intrinsic into
proper SPIR-V/DXIL.

Related to #99159
2025-06-10 14:25:09 +02:00
Helena Kotas
27675ccdd6
[DirectX] Add resource name argument to llvm.dx.handlefrom[implicit]binding intrinsics (#139991)
Adds resource name argument to `llvm.dx.handlefrombinding` and `llvm.dx.handlefromimplicitbinding` intrinsics.
SPIR-V currently does not seem to need the resource names so this change only affects DirectX binding intrinsics.

Part 2/4 of https://github.com/llvm/llvm-project/issues/105059
2025-05-27 22:57:01 -07:00
Sarah Spall
5ba57a81f2
[HLSL] Update Sema Checking Diagnostics for builtins (#138429)
Update how Sema Checking is done for HLSL builtins to allow for better
error messages, mainly using 'err_builtin_invalid_arg_type'.
Try to follow the formula outlined in issue #134721 
Closes #134721
2025-05-21 13:03:07 -07:00
Justin Bogner
f695c8d529
[DirectX][SPIRV] Fix the lowering of dot4add (#140315)
There were some issues with these ops:
- The overload wasn't being specified (`dx.op.dot4AddPacked` vs
`dx.op.dot4AddPacked.i32`)
- The versioning wasn't correct (These ops were added in SM 6.4)
- The argument order was off - while the HLSL function has the
accumulator as the last argument, the DXIL op lists it first.

This fixes the DXIL.td definition and adjusts the LLVM DX and SPIRV
intrinsics to match the argument order in DXIL rather than the argument
order in HLSL.

Fixes #139018
2025-05-17 10:38:24 -07:00
Helena Kotas
520773b47e
[HLSL] Add resource constructor with implicit binding for global resources (#138976)
Adds constructor for resources with implicit binding and applies it to
all resources without binding at the global scope.
Adds Clang builtin function
`__builtin_hlsl_resource_handlefromimplicitbinding` that gets translated
to `llvm.dx|spv.resource.handlefromimplicitbinding` intrinsic calls.
Specific bindings are assigned in DXILResourceImplicitBinding pass.

Design proposals:

https://github.com/llvm/wg-hlsl/blob/main/proposals/0024-implicit-resource-binding.md

https://github.com/llvm/wg-hlsl/blob/main/proposals/0025-resource-constructors.md

One change from the proposals is that the `orderId` parameter is added
onto the constructor. Originally it was supposed to be generated in
codegen when the `llvm.dx|spv.resource.handlefromimplicitbinding` call
is emitted, but that is not possible because the call is inside a
constructor, and the constructor body is generated once per resource
type and not resource instance. So the only way to inject instance-based
data like `orderId` into the
`llvm.dx|spv.resource.handlefromimplicitbinding` call is that it must
come in via the constructor argument.

Closes #136784
2025-05-14 18:41:17 -07:00
Helena Kotas
f1750300aa
[HLSL] Resource initialization by constructors (#135120)
- Adds resource constructor that takes explicit binding for all resource
classes.
- Updates implementation of default resource constructor to initialize
resource handle to `poison`.
- Removes initialization of resource classes from Codegen.
- Initialization of `cbuffer` still needs to happen in `CGHLSLRuntime`
because it does not have a corresponding resource class type.
- Adds `ImplicitCastExpr` for builtin function calls. Sema adds these
automatically when a method of a template class is instantiated, but
some resource classes like `ByteAddressBuffer` are not templates so they
need to have this added explicitly.

Design proposal: llvm/wg-hlsl#197

Closes #134154
2025-04-29 18:42:22 -07:00
Steven Perron
c073c22865
[HLSL] Use hlsl_device address space for getpointer. (#127675)
We add the hlsl_device address space to represent the device memory
space as defined in section 1.7.1.3 of the [HLSL
spec](https://microsoft.github.io/hlsl-specs/specs/hlsl.pdf).

Fixes https://github.com/llvm/llvm-project/issues/127075
2025-04-22 13:26:32 -04:00
Farzon Lotfi
16c84c4475
[DirectX] Add target builtins (#134439)
- fixes #132303
- Moves dot2add from a language builtin to a target builtin.
-  Sets the scaffolding for Sema checks for DX builtins
-  Setup DirectX backend as able to have target builtins
- Adds a DX TargetBuiltins emitter in
`clang/lib/CodeGen/TargetBuiltins/DirectX.cpp`
2025-04-07 12:06:57 -04:00
NAKAMURA Takumi
4088c70f4e CGHLSLBuiltins.cpp: Suppress a warning in #131237 [-Wunused-variable] 2025-04-04 11:05:46 +09:00
Sumit Agarwal
996cf5dc67
[HLSL] Implement dot2add intrinsic (#131237)
Resolves #99221 
Key points: For SPIRV backend, it decompose into a `dot` followed a
`add`.

- [x] Implement dot2add clang builtin,
- [x] Link dot2add clang builtin with hlsl_intrinsics.h
- [x] Add sema checks for dot2add to CheckHLSLBuiltinFunctionCall in
SemaHLSL.cpp
- [x] Add codegen for dot2add to EmitHLSLBuiltinExpr in CGBuiltin.cpp
- [x] Add codegen tests to clang/test/CodeGenHLSL/builtins/dot2add.hlsl
- [x] Add sema tests to clang/test/SemaHLSL/BuiltIns/dot2add-errors.hlsl
- [x] Create the int_dx_dot2add intrinsic in IntrinsicsDirectX.td
- [x] Create the DXILOpMapping of int_dx_dot2add to 162 in DXIL.td
- [x] Create the dot2add.ll and dot2add_errors.ll tests in
llvm/test/CodeGen/DirectX/
2025-04-03 16:23:09 -06:00
Sarah Spall
60efed3f20
[HLSL] Update __builtin_hlsl_dot builtin Sema Checking to fix error when passed an array literal 1u.xxxx (#133941)
update dot builtin sema checking and codegen
new test 
fix tests
Closes #133659
2025-04-02 12:27:01 -07:00
Rahul Joshi
74b7abf154
[IRBuilder] Add new overload for CreateIntrinsic (#131942)
Add a new `CreateIntrinsic` overload with no `Types`, useful for
creating calls to non-overloaded intrinsics that don't need additional
mangling.
2025-03-31 08:10:34 -07:00
Farzon Lotfi
59d06071e9
[NFC][HLSL] Move emitter out of AMDGPU.cpp (#133251)
- Move all HLSL code out of AMDGPU.cpp to CGHLSLBuiltins.cpp
- Fixes accidental reorganization of HLSL code into AMDGPU caused by
(https://github.com/llvm/llvm-project/pull/132252,
https://github.com/llvm/llvm-project/commit/7f920e2e5f70b)
2025-03-27 11:47:14 -04:00