NVAPI improvements (#1512)

* 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.

2 files changed, 9 insertions, 14 deletions

diff --git a/tests/slang-extension/atomic-float-byte-address-buffer.slang b/tests/slang-extension/atomic-float-byte-address-buffer.slang
index a332ddeac..603b92d65 100644
--- a/tests/slang-extension/atomic-float-byte-address-buffer.slang
+++ b/tests/slang-extension/atomic-float-byte-address-buffer.slang
@@ -1,18 +1,14 @@
 // atomic-float-byte-address-buffer.slang
 
 //DISABLE_TEST(compute):COMPARE_COMPUTE_EX:-cpu -compute -output-using-type
-// Produces a different result
-//TEST(compute):COMPARE_COMPUTE_EX:-dx11 -slang -compute -render-features atomic-float -output-using-type -nvapi-register u0
-// Disabled because crashes currently on vulkan 
-// https://vulkan.gpuinfo.org/listdevicescoverage.php?extension=VK_EXT_shader_atomic_float
+//TEST(compute):COMPARE_COMPUTE_EX:-dx11 -slang -compute -render-features atomic-float -output-using-type -nvapi-slot u0
 //TEST(compute, vulkan):COMPARE_COMPUTE_EX:-vk -compute -render-features atomic-float -output-using-type
-// Doesn't work on D3D12 for unknown reasons
-//DISABLE_TEST(compute):COMPARE_COMPUTE_EX:-d3d12 -compute -render-features atomic-float -output-using-type -compile-arg -O2 -nvapi-register u0
-//DISABLE_TEST(compute):COMPARE_COMPUTE_EX:-d3d12 -compute -use-dxil -render-features atomic-float -output-using-type -compile-arg -O2 -nvapi-register u0
+//TEST(compute):COMPARE_COMPUTE_EX:-d3d12 -compute -render-features atomic-float -output-using-type -compile-arg -O2 -nvapi-slot u0
+//TEST(compute):COMPARE_COMPUTE_EX:-d3d12 -compute -use-dxil -render-features atomic-float -output-using-type -compile-arg -O2 -nvapi-slot u0
 //TEST(compute):COMPARE_COMPUTE_EX:-cuda -compute -output-using-type 
 
-// We don't use this, but having this defined makes the 0 slot available if nvapi is going to be used
-// Only strictly necessary on the dx12 path
+// 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 D3D11/D3D12 paths
 //TEST_INPUT:ubuffer(data=[0 0 0 0 ], stride=4):name=nvapiBuffer
 RWStructuredBuffer<int> nvapiBuffer;
 
diff --git a/tests/slang-extension/atomic-int64-byte-address-buffer.slang b/tests/slang-extension/atomic-int64-byte-address-buffer.slang
index 216e55dc2..628c675a2 100644
--- a/tests/slang-extension/atomic-int64-byte-address-buffer.slang
+++ b/tests/slang-extension/atomic-int64-byte-address-buffer.slang
@@ -4,14 +4,13 @@
 //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-register u0
-// Doesn't work on current version of driver. The atomics have no effect perhaps meaning they aren't being decoded correctly
-//DISABLE_TEST(compute):COMPARE_COMPUTE_EX:-slang -compute -dx12 -profile cs_6_0 -use-dxil -render-features atomic-int64 -nvapi-register u0 -compile-arg -O2
+//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
 
-// We don't use this, but having this defined makes the 0 slot available if nvapi is going to be used
-// Only strictly necessary on the dx12 path
+// 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;