Fix for KernelContext threading issue for C++ targets (#1843)

* #include an absolute path didn't work - because paths were taken to always be relative.

* Fix for issue where threading KernelContext was not working on C++ test when there were multiple invocations.

* Improve test for context threading.

1 files changed, 49 insertions, 0 deletions

diff --git a/tests/compute/kernel-context-threading.slang b/tests/compute/kernel-context-threading.slang
new file mode 100644
index 000000000..5cfb171a3
--- /dev/null
+++ b/tests/compute/kernel-context-threading.slang
@@ -0,0 +1,49 @@
+// kernel-context-threading.slang
+
+// This test tests out the slang-ir-explicit-global-context functionality for C++ like targets. 
+// In particular these require that the KernelContext is threaded through functions that access globals. 
+// Currently this is only really applicable to C++, but for completeness all targets are tested.
+
+//TEST(compute):COMPARE_COMPUTE_EX:-cpu -compute -output-using-type -compile-arg -O3 -xslang -matrix-layout-row-major -shaderobj
+//TEST(compute):COMPARE_COMPUTE_EX:-cuda -compute -output-using-type -compile-arg -O3 -xslang -matrix-layout-row-major -shaderobj
+//TEST(compute):COMPARE_COMPUTE_EX:-cuda -compute -output-using-type -compile-arg -O3 -xslang -matrix-layout-row-major -shaderobj
+//TEST(compute):COMPARE_COMPUTE_EX:-slang -compute -output-using-type -xslang -matrix-layout-row-major -shaderobj
+//TEST(compute):COMPARE_COMPUTE_EX:-slang -compute -output-using-type -dx12 -xslang -matrix-layout-row-major -shaderobj
+//TEST(compute, vulkan):COMPARE_COMPUTE_EX:-vk -compute -output-using-type -xslang -matrix-layout-row-major -shaderobj
+
+//TEST_INPUT:cbuffer(data=[1.0 0.0 0.0 0.0  0.0 1.0 0.0 0.0  0.0 0.0 1.0 0.0 10.0 20.0 30.0 1.0]):name matrixBuffer
+ConstantBuffer<float4x4> matrixBuffer;
+
+//TEST_INPUT:ubuffer(data=[0 0 0 0  0 0 0 0  0 0 0 0  0 0 0 0], stride=4):out,name rowOrderMatrixOutput
+RWStructuredBuffer<float> rowOrderMatrixOutput;
+
+void writeRow2(float4 v, int rowIndex)
+{
+    int baseIndex = rowIndex * 4;
+    
+    rowOrderMatrixOutput[baseIndex + 0] = v.x;
+    rowOrderMatrixOutput[baseIndex + 1] = v.y;
+    rowOrderMatrixOutput[baseIndex + 2] = v.z;
+    rowOrderMatrixOutput[baseIndex + 3] = v.w;
+}
+
+// Just to test threading works through multiple levels of functions.
+void writeRow(float4 v, int rowIndex)
+{
+    writeRow2(v, rowIndex);
+}
+
+[numthreads(1, 1, 1)]
+void computeMain(uint3 tid : SV_DispatchThreadID)
+{
+    float4 v = float4(1, 2, 3, 1);
+
+    float4x4 M = matrixBuffer;
+    
+    float4 r = mul(v, M);
+    
+    writeRow(M[0], 0);
+    writeRow(M[1], 1);
+    writeRow(M[2], 2);
+    writeRow(M[3], 3);
+}