Support for float atomics on RWByteAddressBuffer (#1502)

* Fix premake5.lua so it uses the new path needed for OpenCLDebugInfo100.h

* Keep including the includes directory.

* Added the spirv-tools-generated files.

* We don't need to include the spirv/unified1 path because the files needed are actually in the spirv-tools-generated folder.

* Put the build_info.h glslang generated files in external/glslang-generated. Alter premake5.lua to pick up that header.

* First pass at documenting how to build glslang and spirv-tools.

* Improved glsl/spir-v tools README.md

* Added revision.h

* Change how gResources is calculated.
Update about revision.h

* Update docs a little.

* Split out spirv-tools into a separate project for building glslang. This was not necessary on linux, but *is* necessary on windows, because there is a file disassemble.cpp in spirv-tools and in glslang, and this leads to VS choosing only one. With the separate library, the problem is resolved.

* Fix direct-spirv-emit output.

* Update to latest version of spirv headers and spirv-tools.

* Upgrade submodule version of glslang in external.

* Add fPIC to build options of slang-spirv-tools

* WIP adding support for InterlockedAddFp32

* Upgrade slang-binaries to have new glslang.

* Fix issues with Windows slang-glslang binaries, via update of slang-binaries used.

* WIP - atomicAdd. This solution can't work as we can't do (float*) in glsl.

* WIP on atomic float ops.

* Added checking for multiple decls that takes into account __target_intrinsic and __specialized_for_target.
First pass impl of atomic add on float for glsl.

* Split __atomicAdd so extensions are applied appropriately.

* Made Dxc/Fxc support includes.
Use HLSL prelude to pass the path to nvapi
Added -nv-api-path

* Refactor around IncludeHandler and impl of IncludeSystem

* slang-include-handler -> slang-include-system
Have IncludeHandler/Impl defined in slang-preprocessor

* Small comment improvements.

* Document atomic float add addition in target-compatibility.md.

* CUDA float atomic support on RWByteAddressBuffer.

* Add atomic-float-byte-address-buffer-cross.slang

* Removed inappropriate-once.slang - the test is no longer valid when a file is loaded and has a unique identity by default. A test could be made, but would require an API call to create the file (so no unique id).
Improved handling of loadFile - uses uniqueId if has one.

* Work around for testing target overlaps - to avoid exceptions on adding targets.
Simplify PathInfo setup.
Modify single-target-intrinsic.slang - it no longer failed because there were no longer multiple definitions for the same target.

Co-authored-by: Tim Foley <tfoleyNV@users.noreply.github.com>

1 files changed, 62 insertions, 0 deletions

diff --git a/source/slang/hlsl.meta.slang b/source/slang/hlsl.meta.slang
index 62a548555..329a73a33 100644
--- a/source/slang/hlsl.meta.slang
+++ b/source/slang/hlsl.meta.slang
@@ -48,6 +48,29 @@ struct ByteAddressBuffer
     }
 };
 
+
+// Make the GLSL atomicAdd available.
+// We have separate int/float implementations, as the float version requires some specific extensions
+// https://www.khronos.org/registry/OpenGL/extensions/NV/NV_shader_atomic_float.txt
+
+__target_intrinsic(glsl, "atomicAdd($0, $1)")
+__glsl_version(430)
+__glsl_extension(GL_EXT_shader_atomic_float)
+//__glsl_extension(GL_EXT_gpu_shader5)
+float __atomicAdd(__ref float value, float amount);
+
+// Int versions require glsl 4.30
+// https://www.khronos.org/registry/OpenGL-Refpages/gl4/html/atomicAdd.xhtml
+
+__target_intrinsic(glsl, "atomicAdd($0, $1)")
+__glsl_version(430)
+int __atomicAdd(__ref int value, int amount);
+
+__target_intrinsic(glsl, "atomicAdd($0, $1)")
+__glsl_version(430)
+uint __atomicAdd(__ref uint value, uint amount);
+
+
 __intrinsic_op($(kIROp_ByteAddressBufferLoad))
 T __byteAddressBufferLoad<T>(ByteAddressBuffer buffer, int offset);
 
@@ -159,6 +182,41 @@ struct $(item.name)
     {
         return __byteAddressBufferLoad<T>(this, location);
     }
+${{{{
+    if (item.op == kIROp_HLSLRWByteAddressBufferType)
+    {
+}}}}
+
+    // float32 and int64 atomic support. This is a Slang specific extension, it uses
+    // GL_EXT_shader_atomic_float on vk
+    // NvAPI support on DX
+    // NOTE! To use this feature on HLSL, the shader needs to include 'nvHLSLExtns.h' from the NvAPI SDK
+    //
+    __target_intrinsic(hlsl, "($3 = NvInterlockedAddFp32($0, $1, $2))")
+    __target_intrinsic(cuda, "(*$3 = atomicAdd((float*)$0._getPtrAt($1), $2))")
+    void InterlockedAddFp32(uint byteAddress, float valueToAdd, out float originalValue);
+
+    __specialized_for_target(glsl)
+    void InterlockedAddFp32(uint byteAddress, float valueToAdd, out float originalValue)
+    {
+        RWStructuredBuffer<float> buf = __getEquivalentStructuredBuffer<float>(this);
+        originalValue = __atomicAdd(buf[byteAddress / 4], valueToAdd);
+    }
+
+    __target_intrinsic(hlsl, "(NvInterlockedAddFp32($0, $1, $2))")
+    __target_intrinsic(cuda, "atomicAdd((float*)$0._getPtrAt($1), $2)")
+    void InterlockedAddFp32(uint byteAddress, float valueToAdd);
+
+    __specialized_for_target(glsl)
+    void InterlockedAddFp32(uint byteAddress, float valueToAdd)
+    {
+        RWStructuredBuffer<float> buf = __getEquivalentStructuredBuffer<float>(this);
+        __atomicAdd(buf[byteAddress / 4], valueToAdd);
+    }
+
+${{{{
+    }
+}}}}
 
     // Added operations:
 
@@ -1091,6 +1149,10 @@ T dot(vector<T, N> x, vector<T, N> y)
 
 __generic<T : __BuiltinFloatingPointType> vector<T,4> dst(vector<T,4> x, vector<T,4> y);
 
+// Given a RWByteAddressBuffer allow it to be interpretted as a RWStructuredBuffer
+__intrinsic_op($(kIROp_GetEquivalentStructuredBuffer))
+RWStructuredBuffer<T> __getEquivalentStructuredBuffer<T>(RWByteAddressBuffer b);
+
 // Error message
 
 // void errorf( string format, ... );