From c74a700681b0be44a74f16b0f9eaad05bba159d2 Mon Sep 17 00:00:00 2001
From: jsmall-nvidia <jsmall@nvidia.com>
Date: Wed, 22 Jan 2020 11:06:20 -0500
Subject: 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.
---
 prelude/slang-cpp-scalar-intrinsics.h | 22 +++++++++++++++++++++-
 prelude/slang-cuda-prelude.h          |  7 +++++++
 2 files changed, 28 insertions(+), 1 deletion(-)

(limited to 'prelude')

diff --git a/prelude/slang-cpp-scalar-intrinsics.h b/prelude/slang-cpp-scalar-intrinsics.h
index 63fe9c926..6c577733d 100644
--- a/prelude/slang-cpp-scalar-intrinsics.h
+++ b/prelude/slang-cpp-scalar-intrinsics.h
@@ -3,6 +3,11 @@
 
 #include "../slang.h"
 
+#if SLANG_PROCESSOR_X86_64 && SLANG_VC
+// If we have visual studio and 64 bit processor, we can assume we have popcnt, and can include x86 intrinsics
+#   include <intrin.h>
+#endif
+
 #ifdef SLANG_PRELUDE_NAMESPACE
 namespace SLANG_PRELUDE_NAMESPACE {
 #endif
@@ -192,7 +197,22 @@ SLANG_FORCE_INLINE double U32_asdouble(uint32_t low, uint32_t hi)
     return u.d;
 }
 
-
+SLANG_FORCE_INLINE uint32_t U32_countbits(uint32_t v)
+{
+#if SLANG_GCC_FAMILY    
+    return __builtin_popcount(v);
+#elif SLANG_PROCESSOR_X86_64 && SLANG_VC
+    return __popcnt(v);
+#else     
+    uint32_t c = 0;
+    while (v)
+    {
+        c++;
+        v &= v - 1;
+    }
+    return c;
+#endif
+}
 
 
 #ifdef SLANG_PRELUDE_NAMESPACE
diff --git a/prelude/slang-cuda-prelude.h b/prelude/slang-cuda-prelude.h
index 8d100b0db..f78814486 100644
--- a/prelude/slang-cuda-prelude.h
+++ b/prelude/slang-cuda-prelude.h
@@ -161,6 +161,13 @@ SLANG_CUDA_CALL double U32_asdouble(uint32_t low, uint32_t hi)
     return u.d;
 }
 
+SLANG_CUDA_CALL uint32_t U32_countbits(uint32_t v)
+{
+    // https://docs.nvidia.com/cuda/cuda-math-api/group__CUDA__MATH__INTRINSIC__INT.html#group__CUDA__MATH__INTRINSIC__INT_1g43c9c7d2b9ebf202ff1ef5769989be46
+    return __popc(v);
+}
+
+
 /* !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! */
 
 
-- 
cgit v1.2.3