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>

1 files changed, 51 insertions, 8 deletions

diff --git a/source/slang/hlsl.meta.slang b/source/slang/hlsl.meta.slang
index 2ac886f61..66d1cb5e6 100644
--- a/source/slang/hlsl.meta.slang
+++ b/source/slang/hlsl.meta.slang
@@ -9539,6 +9539,39 @@ vector<T,N> faceforward(vector<T,N> n, vector<T,N> i, vector<T,N> ng)
     }
 }
 
+// Helper functions for Metal target
+internal int __metal_clz(int value)
+{
+    __target_switch
+    {
+    case metal: __intrinsic_asm "clz";
+    }
+}
+
+internal uint __metal_clz(uint value)
+{
+    __target_switch
+    {
+    case metal: __intrinsic_asm "clz";
+    }
+}
+
+internal int __metal_ctz(int value)
+{
+    __target_switch
+    {
+    case metal: __intrinsic_asm "ctz";
+    }
+}
+
+internal uint __metal_ctz(uint value)
+{
+    __target_switch
+    {
+    case metal: __intrinsic_asm "ctz";
+    }
+}
+
 /// Find first set bit starting at high bit and working down.
 /// @param value The value to find set bits in.
 /// @return The bit index number of the most significant bit,
@@ -9559,7 +9592,14 @@ int firstbithigh(int value)
     case cuda: __intrinsic_asm "$P_firstbithigh($0)";
     case glsl: __intrinsic_asm "findMSB";
     case hlsl: __intrinsic_asm "firstbithigh";
-    case metal: __intrinsic_asm "clz";
+    case metal:
+        {
+            if ((int)value < 0)
+                value = ~value;
+            if (value == 0)
+                return ~0u;
+            return 31 - __metal_clz(value);
+        }
     case spirv: return spirv_asm {
         OpExtInst $$int result glsl450 FindSMsb $value
     };
@@ -9576,7 +9616,6 @@ vector<int, N> firstbithigh(vector<int, N> value)
     {
     case glsl: __intrinsic_asm "findMSB";
     case hlsl: __intrinsic_asm "firstbithigh";
-    case metal: __intrinsic_asm "clz";
     case spirv: return spirv_asm {
         OpExtInst $$vector<int, N> result glsl450 FindSMsb $value
     };
@@ -9596,7 +9635,14 @@ uint firstbithigh(uint value)
     case cuda: __intrinsic_asm "$P_firstbithigh($0)";
     case glsl: __intrinsic_asm "findMSB";
     case hlsl: __intrinsic_asm "firstbithigh";
-    case metal: __intrinsic_asm "clz";
+    case metal:
+        {
+            if ((int)value < 0)
+                value = ~value;
+            if (value == 0)
+                return ~0u;
+            return 31 - __metal_clz(value);
+        }
     case spirv: return spirv_asm {
         OpExtInst $$uint result glsl450 FindUMsb $value
     };
@@ -9613,7 +9659,6 @@ vector<uint,N> firstbithigh(vector<uint,N> value)
     {
     case glsl: __intrinsic_asm "findMSB";
     case hlsl: __intrinsic_asm "firstbithigh";
-    case metal: __intrinsic_asm "clz";
     case spirv: return spirv_asm {
         OpExtInst $$vector<uint,N> result glsl450 FindUMsb $value
     };
@@ -9639,7 +9684,7 @@ int firstbitlow(int value)
     case cuda: __intrinsic_asm "$P_firstbitlow($0)";
     case glsl: __intrinsic_asm "findLSB";
     case hlsl: __intrinsic_asm "firstbitlow";
-    case metal: __intrinsic_asm "ctz";
+    case metal: return (value==0) ? -1 : __metal_ctz(value);
     case spirv: return spirv_asm {
         OpExtInst $$int result glsl450 FindILsb $value
     };
@@ -9656,7 +9701,6 @@ vector<int,N> firstbitlow(vector<int,N> value)
     {
     case glsl: __intrinsic_asm "findLSB";
     case hlsl: __intrinsic_asm "firstbitlow";
-    case metal: __intrinsic_asm "ctz";
     case spirv: return spirv_asm {
         OpExtInst $$vector<int,N> result glsl450 FindILsb $value
     };
@@ -9676,7 +9720,7 @@ uint firstbitlow(uint value)
     case cuda: __intrinsic_asm "$P_firstbitlow($0)";
     case glsl: __intrinsic_asm "findLSB";
     case hlsl: __intrinsic_asm "firstbitlow";
-    case metal: __intrinsic_asm "ctz";
+    case metal: return (value==0) ? -1 : __metal_ctz(value);
     case spirv: return spirv_asm {
         OpExtInst $$uint result glsl450 FindILsb $value
     };
@@ -9693,7 +9737,6 @@ vector<uint,N> firstbitlow(vector<uint,N> value)
     {
     case glsl: __intrinsic_asm "findLSB";
     case hlsl: __intrinsic_asm "firstbitlow";
-    case metal: __intrinsic_asm "ctz";
     case spirv: return spirv_asm {
         OpExtInst $$vector<uint,N> result glsl450 FindILsb $value
     };