RWByteAddressBuffer::InterlockedCompareExchangeU64 (#1513)

* First pass at incorporating nvapi into test harness.

* D3d12 Atomic Float Add via NVAPI working

* Dx12 atomic float appears to work.

* Atomic float add on Dx12.

* Added atomic64 feature addition to vk.
Fix correct output for atomic-float-byte-address.slang

* Disable atomic float failing tests.

* Upgraded VK headers.

* Detect atomic float availability on VK.

* Try to get test working for in64 atomic.

* Made HLSL prelude controlled via the render-test requirements.

* Added -enable-nvapi to premake.

* Fix D3D12Renderer when NVAPI is not available.

* Small improvements to VKRenderer.

* Improve atomic documentation in target-compatibility.md.

* Fixed NVAPI working on D3D12.

* Test for specific NVAPI features.

* Remove requiredFeatures from Renderer::Desc as was ignored. Tried to document more around nvapiExtnSlot.

* Readded requiredFeatures to Renderer::Desc

* Improve comments in the tests.

* Rename Fp32 -> F32
Added cas-int64-byte-address-buffer.slang test

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

5 files changed, 62 insertions, 9 deletions

diff --git a/tests/slang-extension/atomic-float-byte-address-buffer-cross.slang b/tests/slang-extension/atomic-float-byte-address-buffer-cross.slang
index e99494d5f..584dcada1 100644
--- a/tests/slang-extension/atomic-float-byte-address-buffer-cross.slang
+++ b/tests/slang-extension/atomic-float-byte-address-buffer-cross.slang
@@ -19,9 +19,9 @@ void computeMain(uint3 dispatchThreadID : SV_DispatchThreadID)
     const float delta = anotherBuffer[idx & 3];
     
     float previousValue = 0;
-    outputBuffer.InterlockedAddFp32((idx << 2), 1.0f, previousValue);
+    outputBuffer.InterlockedAddF32((idx << 2), 1.0f, previousValue);
     
     // The sum of values in anotherBuffer should also be added
     int anotherIdx = tid >> 2;
-    outputBuffer.InterlockedAddFp32(anotherIdx << 2, delta);
+    outputBuffer.InterlockedAddF32(anotherIdx << 2, delta);
 }
\ No newline at end of file
diff --git a/tests/slang-extension/atomic-float-byte-address-buffer-cross.slang.glsl b/tests/slang-extension/atomic-float-byte-address-buffer-cross.slang.glsl
index 9fd2f18f9..6e21156aa 100644
--- a/tests/slang-extension/atomic-float-byte-address-buffer-cross.slang.glsl
+++ b/tests/slang-extension/atomic-float-byte-address-buffer-cross.slang.glsl
@@ -14,7 +14,7 @@ layout(std430, binding = 0) buffer _S2 {
 } _S3;
 
 #line 18 "tests/slang-extension/atomic-float-byte-address-buffer-cross.slang"
-void RWByteAddressBuffer_InterlockedAddFp32_0(uint _S4, float _S5, out float _S6)
+void RWByteAddressBuffer_InterlockedAddF32_0(uint _S4, float _S5, out float _S6)
 {
     uint _S7 = _S4 / uint(4);
     float _S8 = (atomicAdd((((_S3)._data[(_S7)])), (_S5)));
@@ -22,7 +22,7 @@ void RWByteAddressBuffer_InterlockedAddFp32_0(uint _S4, float _S5, out float _S6
     return;
 }
 
-void RWByteAddressBuffer_InterlockedAddFp32_1(uint _S9, float _S10)
+void RWByteAddressBuffer_InterlockedAddF32_1(uint _S9, float _S10)
 {
     uint _S11 = _S9 / uint(4);
     float _S12 = (atomicAdd((((_S3)._data[(_S11)])), (_S10)));
@@ -43,8 +43,8 @@ layout(local_size_x = 16, local_size_y = 1, local_size_z = 1) in;void main()
 
 #line 21
     float _S14;
-    RWByteAddressBuffer_InterlockedAddFp32_0(_S13, 1.00000000000000000000, _S14);
-    RWByteAddressBuffer_InterlockedAddFp32_1(uint(int(tid_0 >> 2) << 2), delta_0);
+    RWByteAddressBuffer_InterlockedAddF32_0(_S13, 1.00000000000000000000, _S14);
+    RWByteAddressBuffer_InterlockedAddF32_1(uint(int(tid_0 >> 2) << 2), delta_0);
 
 #line 13
     return;
diff --git a/tests/slang-extension/atomic-float-byte-address-buffer.slang b/tests/slang-extension/atomic-float-byte-address-buffer.slang
index 603b92d65..910fefdfe 100644
--- a/tests/slang-extension/atomic-float-byte-address-buffer.slang
+++ b/tests/slang-extension/atomic-float-byte-address-buffer.slang
@@ -30,12 +30,12 @@ void computeMain(uint3 dispatchThreadID : SV_DispatchThreadID)
     //const float delta = anotherBuffer[idx & 3];
     
     float previousValue = 0;
-    workBuffer.InterlockedAddFp32((idx << 2), 1.0f, previousValue);
-    //workBuffer.InterlockedAddFp32((idx ^ 2) << 2, 2.0f + delta);
+    workBuffer.InterlockedAddF32((idx << 2), 1.0f, previousValue);
+    //workBuffer.InterlockedAddF32((idx ^ 2) << 2, 2.0f + delta);
     
     // The sum of values in anotherBuffer should also be added
     //int anotherIdx = tid >> 2;
-    //workBuffer.InterlockedAddFp32(anotherIdx << 2, delta);
+    //workBuffer.InterlockedAddF32(anotherIdx << 2, delta);
     
     GroupMemoryBarrierWithGroupSync();
     
diff --git a/tests/slang-extension/cas-int64-byte-address-buffer.slang b/tests/slang-extension/cas-int64-byte-address-buffer.slang
new file mode 100644
index 000000000..b75a9fa04
--- /dev/null
+++ b/tests/slang-extension/cas-int64-byte-address-buffer.slang
@@ -0,0 +1,45 @@
+// No atomic support on CPU
+//DISABLE_TEST(compute):COMPARE_COMPUTE_EX:-cpu -compute 
+// No support for int64_t on DX11
+//DISABLE_TEST(compute):COMPARE_COMPUTE_EX:-slang -compute
+// No support for int64_t on fxc - we need SM6.0 and dxil
+// https://docs.microsoft.com/en-us/windows/win32/direct3dhlsl/hlsl-shader-model-6-0-features-for-direct3d-12
+//DISABLE_TEST(compute):COMPARE_COMPUTE_EX:-slang -compute -dx12 -nvapi-slot u0
+//TEST(compute):COMPARE_COMPUTE_EX:-slang -compute -dx12 -profile cs_6_0 -use-dxil -render-features atomic-int64 -nvapi-slot u0 -compile-arg -O2
+//TEST(compute, vulkan):COMPARE_COMPUTE_EX:-vk -compute -render-features atomic-int64
+//TEST(compute):COMPARE_COMPUTE_EX:-cuda -compute
+
+// The test doesn't directly use this, but having this defined makes the 0 slot available if NVAPI is going to be used
+// Only strictly necessary on the D3D12 path
+//TEST_INPUT:ubuffer(data=[0 0 0 0 ], stride=4):name=nvapiBuffer
+RWStructuredBuffer<int> nvapiBuffer;
+
+//TEST_INPUT:ubuffer(data=[0 1 2 3 4 5 6 7]):out,name=outputBuffer
+RWByteAddressBuffer outputBuffer;
+
+[numthreads(16, 1, 1)]
+void computeMain(uint3 dispatchThreadID : SV_DispatchThreadID)
+{    
+    uint tid = dispatchThreadID.x;
+    int idx = (tid & 3) ^ (tid >> 2); 
+
+    // Try directly reading
+    uint2 currentValue2 = outputBuffer.Load2(idx << 8);
+    uint64_t currentValue = uint64_t(currentValue2.y) | currentValue2.x;    
+    
+    while (true)
+    {
+        // This is probably not a great way to do this - InterlockedAddI64 would be better
+        // but we are doing this to test CAS.
+        
+        uint64_t readValue;
+        outputBuffer.InterlockedCompareExchangeU64(idx << 3, currentValue, currentValue + 1, readValue);
+        
+        if (readValue == currentValue)
+        {
+            break;
+        }
+        
+        currentValue = readValue;
+    }
+}
\ No newline at end of file
diff --git a/tests/slang-extension/cas-int64-byte-address-buffer.slang.expected.txt b/tests/slang-extension/cas-int64-byte-address-buffer.slang.expected.txt
new file mode 100644
index 000000000..78d24c356
--- /dev/null
+++ b/tests/slang-extension/cas-int64-byte-address-buffer.slang.expected.txt
@@ -0,0 +1,8 @@
+4
+1
+6
+3
+8
+5
+A
+7