Slang -> CUDA kernel runs correctly in test infrastructure (#1167)

* First pass at BindLocation.

* Added BindSet::init - for initializing with two input constant buffers. Needs better name, and perhaps should be another class.

* Fix handling of constant buffer stripping.
Improved initialization.

* Trying to generalize BindLocation a little more.
Split out CPULikeBindRoot.

* More work to make BindLocation et al work with non uniform bindings.

* Added parsing to a location.

* WIP: Trying to get CPU working with BindLocation.

* Describe problem of knowing the type of the reference point in the binding table.

* More ideas on getBindings fix.

* Remove BindSet as member of BindLocation.

* Added BindLocation::Invalid

* Made BindLocation able to be key in hash

* Use BindLocation for bindings on BindingSet.

* Added cuda and nvrtc categories to test infrastructure.
Disabled CUDA synthetic tests by default.
Fixed such that all tests now produce something in BindLocation style.

* Use m_userIndex instead of m_userData on Resource.
Move the binding setup out of cpu-compute-util (as no longer CPU specific)

* Removed CPUBinding - used BindLocation/BindSet instead.
Fixed some bugs around indexOf around uniform indirection.

* Renamed BindSet::Resource -> BindSet::Value.

* Document BindLocation.

* Fixes for Clang/GCC
Improve invariant requirement handling when constructing from BindPoints.

* WIP: First attempt to run CUDA kernel.

* Fix some issues around doing CUDA kernel launch.

* Fix issues around use of cudaMemCpy .

* Better cuda runtime error checking mechanism.

* Fixed bug in passing parameters to cuda kernel launch.
Simplified initialisation of context.

* WIP: Fix CUDA runtime issues.

* Add explicit CUDA synchronize so failures don't appear on implicit ones.

* Fix problem emitting non shared variable on CUDA.

* Fix some typos in CUDA layout.
Use just a pointer for now for CUDA StucturedBuffer.

* Arg order for CUDA launch was wrong.

* First compute kernel runs on CUDA.

3 files changed, 52 insertions, 6 deletions

diff --git a/source/slang/slang-emit-cuda.cpp b/source/slang/slang-emit-cuda.cpp
index 12807e9e2..39a25aafa 100644
--- a/source/slang/slang-emit-cuda.cpp
+++ b/source/slang/slang-emit-cuda.cpp
@@ -509,7 +509,7 @@ void CUDASourceEmitter::emitModuleImpl(IRModule* module)
         // Output all the thread locals 
         for (auto action : actions)
         {
-            if (action.level == EmitAction::Level::Definition && _isVariable(action.inst->op))
+            if (action.level == EmitAction::Level::Definition && action.inst->op == kIROp_GlobalVar)
             {
                 emitGlobalInst(action.inst);
             }
diff --git a/source/slang/slang-ir-entry-point-uniforms.cpp b/source/slang/slang-ir-entry-point-uniforms.cpp
index ad535b747..388a7004d 100644
--- a/source/slang/slang-ir-entry-point-uniforms.cpp
+++ b/source/slang/slang-ir-entry-point-uniforms.cpp
@@ -452,6 +452,8 @@ void moveEntryPointUniformParamsToGlobalScope(
         case CodeGenTarget::Executable:
         case CodeGenTarget::SharedLibrary:
         case CodeGenTarget::HostCallable:
+        case CodeGenTarget::CUDASource:
+        case CodeGenTarget::PTX:
         {
             context.targetNeedsConstantBuffer = false;
             break;
diff --git a/source/slang/slang-type-layout.cpp b/source/slang/slang-type-layout.cpp
index 772686163..644f54a95 100644
--- a/source/slang/slang-type-layout.cpp
+++ b/source/slang/slang-type-layout.cpp
@@ -730,11 +730,55 @@ struct CPUObjectLayoutRulesImpl : ObjectLayoutRulesImpl
 };
 
 
-// TODO(JS): Most likely wrong! Use CPU layout for CUDA for now
+// TODO(JS): Most likely wrong! Assumes largely CPU layout which is probably not right
 struct CUDAObjectLayoutRulesImpl : CPUObjectLayoutRulesImpl
 {
     typedef CPUObjectLayoutRulesImpl Super;
 
+    virtual SimpleLayoutInfo GetObjectLayout(ShaderParameterKind kind) override
+    {
+        switch (kind)
+        {
+            case ShaderParameterKind::ConstantBuffer:
+                // It's a pointer to the actual uniform data
+                return SimpleLayoutInfo(LayoutResourceKind::Uniform, sizeof(void*), sizeof(void*));
+
+            case ShaderParameterKind::MutableTexture:
+            case ShaderParameterKind::TextureUniformBuffer:
+            case ShaderParameterKind::Texture:
+                // It's a pointer to a texture interface 
+                return SimpleLayoutInfo(LayoutResourceKind::Uniform, sizeof(void*), sizeof(void*));
+
+            case ShaderParameterKind::StructuredBuffer:
+            case ShaderParameterKind::MutableStructuredBuffer:
+                // TODO(JS): We are just storing as a pointer for now
+                // It's a ptr and a size of the amount of elements
+                return SimpleLayoutInfo(LayoutResourceKind::Uniform, sizeof(void*), sizeof(void*));
+
+            case ShaderParameterKind::RawBuffer:
+            case ShaderParameterKind::Buffer:
+            case ShaderParameterKind::MutableRawBuffer:
+            case ShaderParameterKind::MutableBuffer:
+
+                // TODO(JS): We are storing as a pointer for now
+
+                // It's a pointer and a size in bytes
+                return SimpleLayoutInfo(LayoutResourceKind::Uniform, sizeof(void*), sizeof(void*));
+
+            case ShaderParameterKind::SamplerState:
+                // It's a pointer
+                return SimpleLayoutInfo(LayoutResourceKind::Uniform, sizeof(void*), sizeof(void*));
+
+            case ShaderParameterKind::TextureSampler:
+            case ShaderParameterKind::MutableTextureSampler:
+            case ShaderParameterKind::InputRenderTarget:
+                // TODO: how to handle these?
+            default:
+                SLANG_UNEXPECTED("unhandled shader parameter kind");
+                UNREACHABLE_RETURN(SimpleLayoutInfo());
+        }
+    }
+
 };
 
 static CPUObjectLayoutRulesImpl kCPUObjectLayoutRulesImpl;
@@ -747,10 +791,10 @@ LayoutRulesImpl kCPULayoutRulesImpl_ = {
 // CUDA
 
 static CUDAObjectLayoutRulesImpl kCUDAObjectLayoutRulesImpl;
-static CUDALayoutRulesImpl kCUALayoutRulesImpl;
+static CUDALayoutRulesImpl kCUDALayoutRulesImpl;
 
 LayoutRulesImpl kCUDALayoutRulesImpl_ = {
-    &kCPULayoutRulesFamilyImpl, &kCUALayoutRulesImpl, &kCUDAObjectLayoutRulesImpl,
+    &kCUDALayoutRulesFamilyImpl, &kCUDALayoutRulesImpl, &kCUDAObjectLayoutRulesImpl,
 };
 
 
@@ -1033,12 +1077,12 @@ LayoutRulesImpl* CPULayoutRulesFamilyImpl::getStructuredBufferRules()
 
 LayoutRulesImpl* CUDALayoutRulesFamilyImpl::getConstantBufferRules()
 {
-    return &kCPULayoutRulesImpl_;
+    return &kCUDALayoutRulesImpl_;
 }
 
 LayoutRulesImpl* CUDALayoutRulesFamilyImpl::getPushConstantBufferRules()
 {
-    return &kCPULayoutRulesImpl_;
+    return &kCUDALayoutRulesImpl_;
 }
 
 LayoutRulesImpl* CUDALayoutRulesFamilyImpl::getTextureBufferRules()