Updated support to enable batch3 (#8219)

Enable CUDA support for batch 3 tests
    - Enhanced wave operations with exclusive support
    - Added proper identity values for min/max operations
    - Fixed intrinsic name mapping issues
    - Updated test configurations

Co-authored-by: Ellie Hermaszewska <ellieh@nvidia.com>

13 files changed, 17 insertions, 13 deletions

diff --git a/tests/autodiff-dstdlib/determinant.slang b/tests/autodiff-dstdlib/determinant.slang
index d2e699551..f73d3cdfa 100644
--- a/tests/autodiff-dstdlib/determinant.slang
+++ b/tests/autodiff-dstdlib/determinant.slang
@@ -1,5 +1,6 @@
 //TEST(compute, vulkan):COMPARE_COMPUTE_EX:-vk -compute -shaderobj -output-using-type
 //TEST(compute):COMPARE_COMPUTE_EX:-slang -compute -shaderobj -output-using-type
+//TEST(compute):COMPARE_COMPUTE_EX:-slang -compute -shaderobj -output-using-type -cuda
 
 //TEST_INPUT:ubuffer(data=[0 0 0 0], stride=4):out,name=outputBuffer
 RWStructuredBuffer<float> outputBuffer;
diff --git a/tests/hlsl-intrinsic/byte-address-buffer/byte-address-half-atomics.slang b/tests/hlsl-intrinsic/byte-address-buffer/byte-address-half-atomics.slang
index f53d38d74..d23a675b2 100644
--- a/tests/hlsl-intrinsic/byte-address-buffer/byte-address-half-atomics.slang
+++ b/tests/hlsl-intrinsic/byte-address-buffer/byte-address-half-atomics.slang
@@ -3,6 +3,7 @@
 
 // Disabled because validation layer doesn't like vector atomics, although nv driver does allow it.
 //DISABLED_TEST(compute, vulkan):COMPARE_COMPUTE(filecheck-buffer=CHECK): -vk -compute -profile cs_6_2 -render-features half -shaderobj -emit-spirv-directly -output-using-type
+//TEST(compute):COMPARE_COMPUTE(filecheck-buffer=CHECK): -cuda -compute -profile cs_6_2 -render-features half -shaderobj -emit-spirv-directly -output-using-type
 //TEST:SIMPLE(filecheck=SPIRV):-target spirv -entry computeMain -stage compute -emit-spirv-directly -skip-spirv-validation
 //TEST:SIMPLE(filecheck=HLSL):-target hlsl -entry computeMain -profile cs_6_3
 //TEST_INPUT:set tmpBuffer = ubuffer(data=[0 0 0 0], stride=4)
diff --git a/tests/hlsl-intrinsic/debug-break.slang b/tests/hlsl-intrinsic/debug-break.slang
index c16139f72..8fd4567dc 100644
--- a/tests/hlsl-intrinsic/debug-break.slang
+++ b/tests/hlsl-intrinsic/debug-break.slang
@@ -1,7 +1,7 @@
 //TEST:SIMPLE(filecheck=CHECK_GLSL):-stage compute -entry computeMain -target glsl -line-directive-mode none
 // We can't enable because output source includes path to prelude.
 //DISABLE_TEST:SIMPLE:-stage compute -entry computeMain -target cpp -line-directive-mode none
-//DISABLE_TEST:SIMPLE:-stage compute -entry computeMain -target cuda -line-directive-mode none
+//TEST:SIMPLE(filecheck=CHECK_CUDA):-stage compute -entry computeMain -target cuda -line-directive-mode none
 // Not currently supported on HLSL
 //DISABLE_TEST:SIMPLE:-stage compute -entry computeMain -target hlsl -line-directive-mode none
 // With `slang-llvm` this will crash, but the call stack isn't really usable.
@@ -15,6 +15,7 @@
 // CHECK_GLSL: void main
 // CHECK_GLSL: glslDebugBreak{{.*}}();
 
+// CHECK_CUDA:__brkpt()
 RWStructuredBuffer<int> outputBuffer;
 
 [numthreads(4, 1, 1)]
@@ -25,4 +26,4 @@ void computeMain(uint3 dispatchThreadID : SV_DispatchThreadID)
     debugBreak();
 
     outputBuffer[idx] = idx;    
-}
\ No newline at end of file
+}
diff --git a/tests/hlsl-intrinsic/scalar-double-vk-intrinsic.slang b/tests/hlsl-intrinsic/scalar-double-vk-intrinsic.slang
index 128c1c1cd..042c0b62e 100644
--- a/tests/hlsl-intrinsic/scalar-double-vk-intrinsic.slang
+++ b/tests/hlsl-intrinsic/scalar-double-vk-intrinsic.slang
@@ -3,8 +3,7 @@
 //TEST(compute):COMPARE_COMPUTE_EX:-cpu -compute -output-using-type -shaderobj
 //TEST(compute, vulkan):COMPARE_COMPUTE_EX:-vk -compute -output-using-type -shaderobj -render-feature double
 
-// We don't want to run a cuda test here...
-//DISABLE_TEST(compute):COMPARE_COMPUTE_EX:-cuda -compute -shaderobj
+//TEST(compute):COMPARE_COMPUTE_EX:-cuda -compute -shaderobj -output-using-type
 
 //TEST_INPUT:ubuffer(data=[0 0 0 0 0 0 0 0], stride=4):out,name outputBuffer
 RWStructuredBuffer<double> outputBuffer;
@@ -54,4 +53,4 @@ void computeMain(uint3 dispatchThreadID : SV_DispatchThreadID)
     }
     
     outputBuffer[idx] = ft;
-}
\ No newline at end of file
+}
diff --git a/tests/hlsl-intrinsic/wave-active-count-bits.slang b/tests/hlsl-intrinsic/wave-active-count-bits.slang
index 105b95a6f..a7aa48687 100644
--- a/tests/hlsl-intrinsic/wave-active-count-bits.slang
+++ b/tests/hlsl-intrinsic/wave-active-count-bits.slang
@@ -2,6 +2,7 @@
 //DISABLE_TEST:COMPARE_COMPUTE_EX:-cpu -compute -shaderobj
 //DISABLE_TEST:COMPARE_COMPUTE_EX:-slang -compute -shaderobj
 //TEST:COMPARE_COMPUTE_EX:-slang -compute -dx12 -use-dxil -profile cs_6_0 -shaderobj -render-feature hardware-device
+//TEST:COMPARE_COMPUTE_EX:-slang -compute -cuda -profile cs_6_0 -shaderobj -render-feature hardware-device
 //TEST(vulkan):COMPARE_COMPUTE_EX:-vk -compute -shaderobj -render-feature hardware-device
 //TEST:COMPARE_COMPUTE_EX:-cuda -compute -render-features cuda_sm_7_0 -shaderobj
 
@@ -14,4 +15,4 @@ void computeMain(uint3 dispatchThreadID : SV_DispatchThreadID)
     int idx = int(dispatchThreadID.x);
     
     outputBuffer[idx] = int(WaveActiveCountBits(bool(idx & 5)));
-}
\ No newline at end of file
+}
diff --git a/tests/hlsl-intrinsic/wave-broadcast-lane-at-vk.slang b/tests/hlsl-intrinsic/wave-broadcast-lane-at-vk.slang
index b0cff08a9..4a32ab736 100644
--- a/tests/hlsl-intrinsic/wave-broadcast-lane-at-vk.slang
+++ b/tests/hlsl-intrinsic/wave-broadcast-lane-at-vk.slang
@@ -1,5 +1,6 @@
 //TEST_CATEGORY(wave, compute)
 //TEST:COMPARE_COMPUTE_EX:-slang -compute -dx12 -use-dxil -profile cs_6_0 -shaderobj
+//TEST:COMPARE_COMPUTE_EX:-slang -compute -cuda -profile cs_6_0 -shaderobj
 //TEST(vulkan):COMPARE_COMPUTE_EX:-vk -compute -shaderobj
 //TEST:COMPARE_COMPUTE_EX:-wgpu -compute -shaderobj
 //TEST:COMPARE_COMPUTE_EX:-metal -compute -shaderobj
diff --git a/tests/hlsl-intrinsic/wave-is-first-lane.slang b/tests/hlsl-intrinsic/wave-is-first-lane.slang
index 220a5758b..ae986c7b3 100644
--- a/tests/hlsl-intrinsic/wave-is-first-lane.slang
+++ b/tests/hlsl-intrinsic/wave-is-first-lane.slang
@@ -3,7 +3,7 @@
 //DISABLE_TEST:COMPARE_COMPUTE_EX:-slang -compute -shaderobj
 //TEST:COMPARE_COMPUTE_EX:-slang -compute -dx12 -use-dxil -profile cs_6_0 -shaderobj -render-feature hardware-device
 //TEST(vulkan):COMPARE_COMPUTE_EX:-vk -compute -shaderobj -render-feature hardware-device
-//TEST:COMPARE_COMPUTE_EX:-cuda -compute -render-features cuda_sm_7_0 -shaderobj
+//TEST:COMPARE_COMPUTE_EX:-cuda -compute -capability cuda_sm_7_0 -shaderobj
 //TEST:COMPARE_COMPUTE_EX:-wgpu -compute -shaderobj
 //TEST:COMPARE_COMPUTE_EX:-metal -compute -shaderobj
 
diff --git a/tests/hlsl-intrinsic/wave-multi/wave-multi-prefix-max.slang b/tests/hlsl-intrinsic/wave-multi/wave-multi-prefix-max.slang
index 654fd6130..0aca11f38 100644
--- a/tests/hlsl-intrinsic/wave-multi/wave-multi-prefix-max.slang
+++ b/tests/hlsl-intrinsic/wave-multi/wave-multi-prefix-max.slang
@@ -1,7 +1,7 @@
 //TEST_CATEGORY(wave, compute)
 //TEST:COMPARE_COMPUTE_EX(filecheck-buffer=CHECK):-vk -compute -shaderobj -emit-spirv-directly
 //TEST:COMPARE_COMPUTE_EX(filecheck-buffer=CHECK):-vk -compute -shaderobj -emit-spirv-via-glsl
-
+//TEST:COMPARE_COMPUTE_EX(filecheck-buffer=CHECK):-cuda -capability cuda_sm_7_0 -compute -shaderobj
 //TEST:COMPARE_COMPUTE_EX(filecheck-buffer=CHECK):-vk -compute -shaderobj -emit-spirv-directly -xslang -DUSE_GLSL_SYNTAX -allow-glsl
 
 //TEST_INPUT:ubuffer(data=[0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0], stride=4):out,name outputBuffer
diff --git a/tests/hlsl-intrinsic/wave-multi/wave-multi-prefix-min.slang b/tests/hlsl-intrinsic/wave-multi/wave-multi-prefix-min.slang
index 68e1e9c05..321b99a0e 100644
--- a/tests/hlsl-intrinsic/wave-multi/wave-multi-prefix-min.slang
+++ b/tests/hlsl-intrinsic/wave-multi/wave-multi-prefix-min.slang
@@ -1,7 +1,7 @@
 //TEST_CATEGORY(wave, compute)
 //TEST:COMPARE_COMPUTE_EX(filecheck-buffer=CHECK):-vk -compute -shaderobj -emit-spirv-directly
 //TEST:COMPARE_COMPUTE_EX(filecheck-buffer=CHECK):-vk -compute -shaderobj -emit-spirv-via-glsl
-
+//TEST:COMPARE_COMPUTE_EX(filecheck-buffer=CHECK):-cuda -capability cuda_sm_7_0 -compute -shaderobj
 //TEST:COMPARE_COMPUTE_EX(filecheck-buffer=CHECK):-vk -compute -shaderobj -emit-spirv-directly -xslang -DUSE_GLSL_SYNTAX -allow-glsl
 
 //TEST_INPUT:ubuffer(data=[0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0], stride=4):out,name outputBuffer
diff --git a/tests/hlsl-intrinsic/wave-multi/wave-multi-prefix-scalar-functional.slang b/tests/hlsl-intrinsic/wave-multi/wave-multi-prefix-scalar-functional.slang
index 5de34b20a..67367f264 100644
--- a/tests/hlsl-intrinsic/wave-multi/wave-multi-prefix-scalar-functional.slang
+++ b/tests/hlsl-intrinsic/wave-multi/wave-multi-prefix-scalar-functional.slang
@@ -4,7 +4,7 @@
 
 //TEST:COMPARE_COMPUTE_EX:-slang -compute -dx12 -use-dxil -profile sm_6_5 -shaderobj
 //TEST:COMPARE_COMPUTE_EX:-vk -compute -shaderobj
-//TEST:COMPARE_COMPUTE_EX:-cuda -compute -render-features cuda_sm_7_0 -shaderobj
+//TEST:COMPARE_COMPUTE_EX:-cuda -compute -capability cuda_sm_7_0 -shaderobj
 
 //TEST_INPUT:ubuffer(data=[0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0], stride=4):out,name outputBuffer
 RWStructuredBuffer<uint> outputBuffer;
diff --git a/tests/hlsl-intrinsic/wave-prefix-product.slang b/tests/hlsl-intrinsic/wave-prefix-product.slang
index 774f5996e..f8c0ed57a 100644
--- a/tests/hlsl-intrinsic/wave-prefix-product.slang
+++ b/tests/hlsl-intrinsic/wave-prefix-product.slang
@@ -3,7 +3,7 @@
 //DISABLE_TEST:COMPARE_COMPUTE_EX:-slang -compute -shaderobj
 //TEST:COMPARE_COMPUTE_EX:-slang -compute -dx12 -use-dxil -profile cs_6_0 -shaderobj -render-feature hardware-device
 //TEST(vulkan):COMPARE_COMPUTE_EX:-vk -compute -shaderobj -render-feature hardware-device
-//TEST:COMPARE_COMPUTE_EX:-cuda -compute -render-features cuda_sm_7_0 -shaderobj
+//TEST:COMPARE_COMPUTE_EX:-cuda -compute -capability cuda_sm_7_0 -shaderobj
 //TEST:COMPARE_COMPUTE_EX:-wgpu -compute -shaderobj
 //TEST:COMPARE_COMPUTE_EX:-metal -compute -shaderobj
 
diff --git a/tests/hlsl-intrinsic/wave-prefix-sum.slang b/tests/hlsl-intrinsic/wave-prefix-sum.slang
index 4f7c2912d..8a092e20d 100644
--- a/tests/hlsl-intrinsic/wave-prefix-sum.slang
+++ b/tests/hlsl-intrinsic/wave-prefix-sum.slang
@@ -3,7 +3,7 @@
 //DISABLE_TEST:COMPARE_COMPUTE_EX:-slang -compute -shaderobj
 //TEST:COMPARE_COMPUTE_EX:-slang -compute -dx12 -use-dxil -profile cs_6_0 -shaderobj -render-feature hardware-device
 //TEST(vulkan):COMPARE_COMPUTE_EX:-vk -compute -shaderobj -render-feature hardware-device
-//TEST:COMPARE_COMPUTE_EX:-cuda -compute -render-features cuda_sm_7_0 -shaderobj
+//TEST:COMPARE_COMPUTE_EX:-cuda -compute -capability cuda_sm_7_0 -shaderobj
 //TEST:COMPARE_COMPUTE_EX:-wgpu -compute -shaderobj
 //TEST:COMPARE_COMPUTE_EX:-metal -compute -shaderobj
 
diff --git a/tests/hlsl-intrinsic/wave-vector.slang b/tests/hlsl-intrinsic/wave-vector.slang
index f786794ec..4c3b7ccb6 100644
--- a/tests/hlsl-intrinsic/wave-vector.slang
+++ b/tests/hlsl-intrinsic/wave-vector.slang
@@ -3,7 +3,7 @@
 //DISABLE_TEST:COMPARE_COMPUTE_EX:-slang -compute -shaderobj
 //TEST:COMPARE_COMPUTE_EX:-slang -compute -dx12 -use-dxil -profile cs_6_0 -shaderobj -render-feature hardware-device
 //TEST(vulkan):COMPARE_COMPUTE_EX:-vk -compute -shaderobj -render-feature hardware-device
-//TEST:COMPARE_COMPUTE_EX:-cuda -compute -render-features cuda_sm_7_0 -shaderobj
+//TEST:COMPARE_COMPUTE_EX:-cuda -compute -capability cuda_sm_7_0 -shaderobj
 //TEST:COMPARE_COMPUTE_EX:-wgpu -compute -shaderobj
 //TEST:COMPARE_COMPUTE_EX:-metal -compute -shaderobj