From 77d59713ac665785b7ebee4ad2b5dcbb73cf5af5 Mon Sep 17 00:00:00 2001
From: jsmall-nvidia <jsmall@nvidia.com>
Date: Tue, 21 Apr 2020 14:09:36 -0400
Subject: Small Improvements around Wave Intrinsics (#1328)

* Fix issues in wave-mask/wave.slang tests.
WaveGetActiveMask -> WaveGetConvergedMask.
Update target-compatibility.md

* First pass at wave-intrinsics.md documentation.
Write up around WaveMaskSharedSync.

* Added more of the Wave intrinsics as WaveMask intrinsics.
Improvements to documentation around wave-intrinsics.

* Add the Wave intrinsics for SM6.5 for WaveMask
Expand WaveMask intrinsics
Improve WaveMask documentation

* Added WaveMaskIsFirstLane.

* Added WaveGetConvergedMask for glsl and hlsl.
Added wave-get-converged-mask.slang test.

* WaveGetActiveMask/Multi and WageGetConvergedMask/Multi

* Improve Wave intrinsics docs.
Adde WaveGetActveMulti WaveGetConvergedMulti, WaveGetActiveMask (for vk/hlsl).

* Enable GLSL WaveMultiPrefixBitAnd.

* Re-add definitions of f16tof32 and f32to16 from #1326

* Remove multiple definition of f32tof16
Disable optix call to Ray trace test, if OPTIX not available.

* Improve wave intrinsics documetnation - remove the __generic as part of definitions, small improvements.

* Change comment to try and trigger build.
---
 .../hlsl-intrinsic/wave-mask/wave-get-active.slang  | 21 +++++++++++++++++++++
 .../wave-mask/wave-get-active.slang.expected.txt    |  8 ++++++++
 .../wave-mask/wave-get-converged.slang              | 18 ++++++++++++++++++
 .../wave-mask/wave-get-converged.slang.expected.txt |  8 ++++++++
 4 files changed, 55 insertions(+)
 create mode 100644 tests/hlsl-intrinsic/wave-mask/wave-get-active.slang
 create mode 100644 tests/hlsl-intrinsic/wave-mask/wave-get-active.slang.expected.txt
 create mode 100644 tests/hlsl-intrinsic/wave-mask/wave-get-converged.slang
 create mode 100644 tests/hlsl-intrinsic/wave-mask/wave-get-converged.slang.expected.txt

(limited to 'tests')

diff --git a/tests/hlsl-intrinsic/wave-mask/wave-get-active.slang b/tests/hlsl-intrinsic/wave-mask/wave-get-active.slang
new file mode 100644
index 000000000..393cf7199
--- /dev/null
+++ b/tests/hlsl-intrinsic/wave-mask/wave-get-active.slang
@@ -0,0 +1,21 @@
+//DISABLE_TEST(compute):COMPARE_COMPUTE_EX:-cpu -compute 
+//DISABLE_TEST(compute):COMPARE_COMPUTE_EX:-slang -compute
+//TEST(compute):COMPARE_COMPUTE_EX:-slang -compute -dx12 -use-dxil -profile cs_6_0
+//TEST(compute, vulkan):COMPARE_COMPUTE_EX:-vk -compute
+//DISABLE_TEST(compute):COMPARE_COMPUTE_EX:-cuda -compute
+
+//TEST_INPUT:ubuffer(data=[0 0 0 0 0 0 0 0], stride=4):out,name outputBuffer
+RWStructuredBuffer<int> outputBuffer;
+
+[numthreads(4, 1, 1)]
+void computeMain(uint3 dispatchThreadID : SV_DispatchThreadID)
+{
+    int idx = int(dispatchThreadID.x);
+    
+    WaveMask mask = WaveGetActiveMask();
+    uint4 multi = WaveGetActiveMulti();
+    
+    // Should be 0xf, because numthreads is 4, 1, 1
+    outputBuffer[idx * 2 + 0] = mask;
+    outputBuffer[idx * 2 + 1] = multi.x;
+}
\ No newline at end of file
diff --git a/tests/hlsl-intrinsic/wave-mask/wave-get-active.slang.expected.txt b/tests/hlsl-intrinsic/wave-mask/wave-get-active.slang.expected.txt
new file mode 100644
index 000000000..e58dced3d
--- /dev/null
+++ b/tests/hlsl-intrinsic/wave-mask/wave-get-active.slang.expected.txt
@@ -0,0 +1,8 @@
+F
+F
+F
+F
+F
+F
+F
+F
diff --git a/tests/hlsl-intrinsic/wave-mask/wave-get-converged.slang b/tests/hlsl-intrinsic/wave-mask/wave-get-converged.slang
new file mode 100644
index 000000000..9b2e814f2
--- /dev/null
+++ b/tests/hlsl-intrinsic/wave-mask/wave-get-converged.slang
@@ -0,0 +1,18 @@
+//DISABLE_TEST(compute):COMPARE_COMPUTE_EX:-cpu -compute 
+//DISABLE_TEST(compute):COMPARE_COMPUTE_EX:-slang -compute
+//TEST(compute):COMPARE_COMPUTE_EX:-slang -compute -dx12 -use-dxil -profile cs_6_0
+//TEST(compute, vulkan):COMPARE_COMPUTE_EX:-vk -compute
+//TEST(compute):COMPARE_COMPUTE_EX:-cuda -compute
+
+//TEST_INPUT:ubuffer(data=[0 0 0 0 0 0 0 0], stride=4):out,name outputBuffer
+RWStructuredBuffer<int> outputBuffer;
+
+[numthreads(4, 1, 1)]
+void computeMain(uint3 dispatchThreadID : SV_DispatchThreadID)
+{
+    int idx = int(dispatchThreadID.x);
+    
+    // Should be 0xf, because numthreads is 4, 1, 1
+    outputBuffer[idx * 2 + 0] = WaveGetConvergedMask();
+    outputBuffer[idx * 2 + 1] = WaveGetConvergedMulti().x;
+}
\ No newline at end of file
diff --git a/tests/hlsl-intrinsic/wave-mask/wave-get-converged.slang.expected.txt b/tests/hlsl-intrinsic/wave-mask/wave-get-converged.slang.expected.txt
new file mode 100644
index 000000000..e58dced3d
--- /dev/null
+++ b/tests/hlsl-intrinsic/wave-mask/wave-get-converged.slang.expected.txt
@@ -0,0 +1,8 @@
+F
+F
+F
+F
+F
+F
+F
+F
-- 
cgit v1.2.3