WIP HLSL intrinsic coverage (#1171)

* Added hlsl-intrinsic test folder.
Enabled ceil as works across targets.

* log10 support.

* Fix float % on CPU/CUDA to match HLSL which is fmod (not fremainder).

* Added log10 tests back to scalar-float.slang

* Don't add the ( for $Sx - it's clearer what's going on without it.

* Works on CUDA/CPU. Problem with asint/asuint do not seem to be found.

* Only asuint exists for double.

* Support countbits on CUDA and C++.

* Fix typo in C++ population count.

* First pass at int vector intrinsic tests.

* Swizzle for int.

* Bit cast tests on CUDA.

* Fix warning on gcc.

* Fix bit-cast-double execution on CUDA.

* scalar-int test working on gcc release.

10 files changed, 116 insertions, 1 deletions

diff --git a/tests/compute/bit-cast-double.slang b/tests/hlsl-intrinsic/bit-cast-double.slang
index 6b0925a58..b52c9de2b 100644
--- a/tests/compute/bit-cast-double.slang
+++ b/tests/hlsl-intrinsic/bit-cast-double.slang
@@ -1,7 +1,8 @@
 //TEST(compute):COMPARE_COMPUTE_EX:-slang -compute -cpu
 //TEST(compute):COMPARE_COMPUTE_EX:-slang -compute
 //TEST(compute):COMPARE_COMPUTE_EX:-slang -compute -dx12
-//TEST(compute, vulkan):COMPARE_COMPUTE_EX:-vk -compute
+//TEST(compute, vulkan):COMPARE_COMPUTE_EX:-slang -vk -compute
+//TEST(compute):COMPARE_COMPUTE_EX:-slang -cuda -compute
 
 //TEST_INPUT:ubuffer(data=[0 0 0 0], stride=4):out,name=outputBuffer
 RWStructuredBuffer<int> outputBuffer;
diff --git a/tests/compute/bit-cast-double.slang.expected.txt b/tests/hlsl-intrinsic/bit-cast-double.slang.expected.txt
index bc856dafa..bc856dafa 100644
--- a/tests/compute/bit-cast-double.slang.expected.txt
+++ b/tests/hlsl-intrinsic/bit-cast-double.slang.expected.txt
diff --git a/tests/compute/bit-cast.slang b/tests/hlsl-intrinsic/bit-cast.slang
index a8fec61e3..4ecd911d7 100644
--- a/tests/compute/bit-cast.slang
+++ b/tests/hlsl-intrinsic/bit-cast.slang
@@ -2,6 +2,7 @@
 //TEST(compute):COMPARE_COMPUTE_EX:-slang -compute
 //TEST(compute):COMPARE_COMPUTE_EX:-slang -compute -dx12
 //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 0 0 0 0], stride=4):out,name outputBuffer
 
diff --git a/tests/compute/bit-cast.slang.expected.txt b/tests/hlsl-intrinsic/bit-cast.slang.expected.txt
index f5414793f..f5414793f 100644
--- a/tests/compute/bit-cast.slang.expected.txt
+++ b/tests/hlsl-intrinsic/bit-cast.slang.expected.txt
diff --git a/tests/hlsl-intrinsic/scalar-int.slang b/tests/hlsl-intrinsic/scalar-int.slang
new file mode 100644
index 000000000..7fccc8645
--- /dev/null
+++ b/tests/hlsl-intrinsic/scalar-int.slang
@@ -0,0 +1,25 @@
+//TEST(compute):COMPARE_COMPUTE_EX:-cpu -compute 
+//TEST(compute):COMPARE_COMPUTE_EX:-slang -compute
+//TEST(compute):COMPARE_COMPUTE_EX:-slang -compute -dx12
+//TEST(compute, vulkan):COMPARE_COMPUTE_EX:-vk -compute
+//TEST(compute, vulkan):COMPARE_COMPUTE_EX:-cuda -compute
+
+//TEST_INPUT:ubuffer(data=[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);
+    
+    int ti =0;
+    
+    ti += max(2, idx);
+    ti += min(idx, 1);
+    ti += abs(idx - 2);
+    ti += (idx * 3) % 5;
+    
+    ti += clamp(idx * 10, 11, 23);
+    
+    outputBuffer[idx] = ti; 
+}
\ No newline at end of file
diff --git a/tests/hlsl-intrinsic/scalar-int.slang.expected.txt b/tests/hlsl-intrinsic/scalar-int.slang.expected.txt
new file mode 100644
index 000000000..4efe33dc6
--- /dev/null
+++ b/tests/hlsl-intrinsic/scalar-int.slang.expected.txt
@@ -0,0 +1,4 @@
+F
+12
+18
+20
diff --git a/tests/hlsl-intrinsic/scalar-uint.slang b/tests/hlsl-intrinsic/scalar-uint.slang
new file mode 100644
index 000000000..eb16bb3b8
--- /dev/null
+++ b/tests/hlsl-intrinsic/scalar-uint.slang
@@ -0,0 +1,26 @@
+//TEST(compute):COMPARE_COMPUTE_EX:-cpu -compute 
+//TEST(compute):COMPARE_COMPUTE_EX:-slang -compute
+//TEST(compute):COMPARE_COMPUTE_EX:-slang -compute -dx12
+//TEST(compute, vulkan):COMPARE_COMPUTE_EX:-vk -compute
+//TEST(compute, vulkan):COMPARE_COMPUTE_EX:-cuda -compute
+
+//TEST_INPUT:ubuffer(data=[0 0 0 0], stride=4):out,name outputBuffer
+RWStructuredBuffer<int> outputBuffer;
+
+[numthreads(4, 1, 1)]
+void computeMain(uint3 dispatchThreadID : SV_DispatchThreadID)
+{
+    uint idx = dispatchThreadID.x;
+    
+    uint ti = 0;
+    
+    ti += max(2, idx);
+    ti += min(idx, 1);
+    ti += (idx * 3) %5;
+    
+    ti += clamp(idx * 10, 11, 23);
+    
+    ti += countbits(idx * 13);
+    
+    outputBuffer[idx] = ti; 
+}
\ No newline at end of file
diff --git a/tests/hlsl-intrinsic/scalar-uint.slang.expected.txt b/tests/hlsl-intrinsic/scalar-uint.slang.expected.txt
new file mode 100644
index 000000000..1a7befbe1
--- /dev/null
+++ b/tests/hlsl-intrinsic/scalar-uint.slang.expected.txt
@@ -0,0 +1,4 @@
+D
+14
+1B
+23
diff --git a/tests/hlsl-intrinsic/vector-int.slang b/tests/hlsl-intrinsic/vector-int.slang
new file mode 100644
index 000000000..9a3f1faaa
--- /dev/null
+++ b/tests/hlsl-intrinsic/vector-int.slang
@@ -0,0 +1,50 @@
+//TEST(compute):COMPARE_COMPUTE_EX:-cpu -compute 
+//TEST(compute):COMPARE_COMPUTE_EX:-slang -compute
+//TEST(compute):COMPARE_COMPUTE_EX:-slang -compute -dx12
+//TEST(compute, vulkan):COMPARE_COMPUTE_EX:-vk -compute
+//TEST(compute, vulkan):COMPARE_COMPUTE_EX:-cuda -compute
+
+//TEST_INPUT:ubuffer(data=[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);
+
+    int3 a = { idx + 10, idx - 9, idx + 3};
+    int3 b = { idx * 2, idx * 3, idx * 10};
+    
+    int3 t = { 0, 0, 0};
+    
+    t += max(a, b);
+    t += min(a, b);
+    t += abs(a);
+    t += b % 5;
+    
+    t += clamp(a, int3(10), int3(23));
+   
+    // Swizzle
+    t += a.zyx;
+    // Swizzle from scalar
+    t += idx.xxx;
+    
+#if 0 
+    // TODO(JS): On C++ not all accesses are turned into getAt/setAt, so disable for now.
+    // Reads
+    {   
+        t += int3(a[idx % 3], a[0], b[2]);
+    }
+   
+    // Writes 
+    {   
+        int3 v = int3(b[(idx + 1) % 3], b[(idx + 2) % 3], b[(idx + 3) % 3]);
+        v[1] += v[0];
+        v[idx & 1] += v[2];
+    
+        t += v;
+    }
+#endif   
+   
+    outputBuffer[idx] = t.x + t.y + t.z;
+}
\ No newline at end of file
diff --git a/tests/hlsl-intrinsic/vector-int.slang.expected.txt b/tests/hlsl-intrinsic/vector-int.slang.expected.txt
new file mode 100644
index 000000000..e4c658883
--- /dev/null
+++ b/tests/hlsl-intrinsic/vector-int.slang.expected.txt
@@ -0,0 +1,4 @@
+3C
+5B
+75
+8F