Fix CUDA backend missing U32_firstbitlow implementation (#7921)

* Initial plan

* Add U32_firstbitlow implementation for CUDA and CPP backends

Co-authored-by: bmillsNV <163073245+bmillsNV@users.noreply.github.com>

* Add I32_firstbitlow and comprehensive testing for signed/unsigned firstbitlow

Co-authored-by: bmillsNV <163073245+bmillsNV@users.noreply.github.com>

* Convert firstbitlow test to use inline filecheck syntax

Co-authored-by: ArielG-NV <159081215+ArielG-NV@users.noreply.github.com>

* Add U32_firstbithigh and I32_firstbithigh implementations for CUDA and CPP backends

Co-authored-by: csyonghe <2652293+csyonghe@users.noreply.github.com>

* Update prelude/slang-cpp-scalar-intrinsics.h

* Update prelude/slang-cpp-scalar-intrinsics.h

* Update prelude/slang-cpp-scalar-intrinsics.h

* Refactor Metal bit intrinsics to handle zero case correctly

Co-authored-by: ArielG-NV <159081215+ArielG-NV@users.noreply.github.com>

* Update slang-cuda-prelude.h

remove fake links

* Update hlsl.meta.slang

* if -1, return -1 due to implicit hlsl rule

* -1 or 0 is ~0u as per hlsl implictly

* 0 or -1 as per hlsl

* fix the math to map to hlsl

* fix compile error

* forgot `31 - clz`

* format code (#7943)

Co-authored-by: slangbot <186143334+slangbot@users.noreply.github.com>

* Update source/slang/hlsl.meta.slang

* Update source/slang/hlsl.meta.slang

* Update source/slang/hlsl.meta.slang

* Update source/slang/hlsl.meta.slang

---------

Co-authored-by: copilot-swe-agent[bot] <198982749+Copilot@users.noreply.github.com>
Co-authored-by: bmillsNV <163073245+bmillsNV@users.noreply.github.com>
Co-authored-by: ArielG-NV <159081215+ArielG-NV@users.noreply.github.com>
Co-authored-by: csyonghe <2652293+csyonghe@users.noreply.github.com>
Co-authored-by: ArielG-NV <aglasroth@nvidia.com>
Co-authored-by: slangbot <ellieh+slangbot@nvidia.com>
Co-authored-by: slangbot <186143334+slangbot@users.noreply.github.com>

2 files changed, 81 insertions, 2 deletions

diff --git a/prelude/slang-cpp-scalar-intrinsics.h b/prelude/slang-cpp-scalar-intrinsics.h
index 9b045941a..731fd02be 100644
--- a/prelude/slang-cpp-scalar-intrinsics.h
+++ b/prelude/slang-cpp-scalar-intrinsics.h
@@ -717,6 +717,50 @@ SLANG_FORCE_INLINE uint32_t U32_countbits(uint32_t v)
 #endif
 }
 
+SLANG_FORCE_INLINE uint32_t U32_firstbitlow(uint32_t v)
+{
+    if (v == 0)
+        return ~0u;
+
+#if SLANG_GCC_FAMILY && !defined(SLANG_LLVM)
+    // __builtin_ctz returns number of trailing zeros, which is the 0-based index of first set bit
+    return __builtin_ctz(v);
+#elif SLANG_PROCESSOR_X86_64 && SLANG_VC
+    // _BitScanForward returns 1 on success, 0 on failure, and sets index
+    unsigned long index;
+    return _BitScanForward(&index, v) ? index : ~0u;
+#else
+    // Generic implementation - find first set bit
+    uint32_t result = 0;
+    while (result < 32 && !(v & (1u << result)))
+        result++;
+    return result;
+#endif
+}
+
+SLANG_FORCE_INLINE uint32_t U32_firstbithigh(uint32_t v)
+{
+    if ((int32_t)v < 0)
+        v = ~v;
+    if (v == 0)
+        return ~0u;
+#if SLANG_GCC_FAMILY && !defined(SLANG_LLVM)
+    // __builtin_clz returns number of leading zeros
+    // firstbithigh should return 0-based bit position of MSB
+    return 31 - __builtin_clz(v);
+#elif SLANG_PROCESSOR_X86_64 && SLANG_VC
+    // _BitScanReverse returns 1 on success, 0 on failure, and sets index
+    unsigned long index;
+    return _BitScanReverse(&index, v) ? index : ~0u;
+#else
+    // Generic implementation - find highest set bit
+    int result = 31;
+    while (result >= 0 && !(v & (1u << result)))
+        result--;
+    return result;
+#endif
+}
+
 // ----------------------------- I32 -----------------------------------------
 
 SLANG_FORCE_INLINE int32_t I32_abs(int32_t f)
@@ -755,6 +799,16 @@ SLANG_FORCE_INLINE uint32_t I32_countbits(int32_t v)
     return U32_countbits(uint32_t(v));
 }
 
+SLANG_FORCE_INLINE uint32_t I32_firstbitlow(int32_t v)
+{
+    return U32_firstbitlow(uint32_t(v));
+}
+
+SLANG_FORCE_INLINE uint32_t I32_firstbithigh(int32_t v)
+{
+    return U32_firstbithigh(uint32_t(v));
+}
+
 // ----------------------------- U64 -----------------------------------------
 
 SLANG_FORCE_INLINE uint64_t U64_abs(uint64_t f)
diff --git a/prelude/slang-cuda-prelude.h b/prelude/slang-cuda-prelude.h
index fd79b77aa..3ebdbe777 100644
--- a/prelude/slang-cuda-prelude.h
+++ b/prelude/slang-cuda-prelude.h
@@ -2081,10 +2081,26 @@ SLANG_FORCE_INLINE SLANG_CUDA_CALL double U32_asdouble(uint32_t low, uint32_t hi
 
 SLANG_FORCE_INLINE 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);
 }
 
+SLANG_FORCE_INLINE SLANG_CUDA_CALL uint32_t U32_firstbitlow(uint32_t v)
+{
+    // __ffs returns 1-based bit position or 0 if no bits set
+    // firstbitlow should return 0-based bit position or ~0u if no bits set
+    return v == 0 ? ~0u : (__ffs(v) - 1);
+}
+
+SLANG_FORCE_INLINE SLANG_CUDA_CALL uint32_t U32_firstbithigh(uint32_t v)
+{
+    // maps to hlsl firstbithigh
+    if ((int32_t)v < 0)
+        v = ~v;
+    if (v == 0)
+        return ~0u;
+    return 31 - __clz(v);
+}
+
 // ----------------------------- I32 -----------------------------------------
 
 // Unary
@@ -2125,6 +2141,16 @@ SLANG_FORCE_INLINE SLANG_CUDA_CALL uint32_t I32_countbits(int32_t v)
     return U32_countbits(uint32_t(v));
 }
 
+SLANG_FORCE_INLINE SLANG_CUDA_CALL uint32_t I32_firstbitlow(int32_t v)
+{
+    return U32_firstbitlow(uint32_t(v));
+}
+
+SLANG_FORCE_INLINE SLANG_CUDA_CALL uint32_t I32_firstbithigh(int32_t v)
+{
+    return U32_firstbithigh(uint32_t(v));
+}
+
 // ----------------------------- U64 -----------------------------------------
 
 SLANG_FORCE_INLINE SLANG_CUDA_CALL int64_t U64_abs(uint64_t f)
@@ -2143,7 +2169,6 @@ SLANG_FORCE_INLINE SLANG_CUDA_CALL int64_t U64_max(uint64_t a, uint64_t b)
 
 SLANG_FORCE_INLINE SLANG_CUDA_CALL uint32_t U64_countbits(uint64_t v)
 {
-    // https://docs.nvidia.com/cuda/cuda-math-api/group__CUDA__MATH__INTRINSIC__INT.html#group__CUDA__MATH__INTRINSIC__INT_1g43c9c7d2b9ebf202ff1ef5769989be46
     return __popcll(v);
 }