Support status argument for GatherXXX

This commit adds an argument to all texture GatherXXX functions. The new argument is for "status" as described in SM5.0 definision. Close shader-slang#4466
jkwak-work · Jul 9, 2024 · df793d1 · df793d1
1 parent ddd14be
commit df793d1
Show file tree

Hide file tree

Showing 2 changed files with 50 additions and 18 deletions.
diff --git a/source/slang/hlsl.meta.slang b/source/slang/hlsl.meta.slang
@@ -2328,11 +2328,17 @@ ${{{{
         {
             auto componentName = componentNames[componentId];
             auto glslComponent = (isCmp ? "" :glslComponentNames[componentId == 0 ? 0 : componentId - 1]);
-    }}}}
+
+            for (bool isStatus : { false, true })
+            {
+                const char* statusDecl = isStatus ? ", out uint status" : "";
+                const char* statusInit = isStatus ? " status = 0;\n" : "";
+}}}}
     [ForceInline]
     [require(glsl_hlsl_metal_spirv, texture_gather)]
-    vector<T,4> Gather$(cmp)$(componentName)($(samplerStateType)$(samplerStateParam) vector<float, Shape.dimensions+isArray> location $(cmpParam))
+    vector<T,4> Gather$(cmp)$(componentName)($(samplerStateType)$(samplerStateParam) vector<float, Shape.dimensions+isArray> location $(cmpParam) $(statusDecl))
     {
+        $(statusInit)
         __target_switch
         {
         case hlsl: __intrinsic_asm ".Gather$(cmp)$(componentName)";
@@ -2345,8 +2351,9 @@ ${{{{
     }
     [ForceInline]
     [require(glsl_hlsl_metal_spirv, texture_gather)]
-    vector<T,4> Gather$(cmp)$(componentName)($(samplerStateType)$(samplerStateParam) vector<float, Shape.dimensions+isArray> location $(cmpParam), constexpr vector<int, Shape.planeDimensions> offset)
+    vector<T,4> Gather$(cmp)$(componentName)($(samplerStateType)$(samplerStateParam) vector<float, Shape.dimensions+isArray> location $(cmpParam), constexpr vector<int, Shape.planeDimensions> offset $(statusDecl))
     {
+        $(statusInit)
         __target_switch
         {
         case hlsl: __intrinsic_asm ".Gather$(cmp)$(componentName)";
@@ -2363,8 +2370,10 @@ ${{{{
         constexpr vector<int, Shape.planeDimensions> offset1,
         constexpr vector<int, Shape.planeDimensions> offset2,
         constexpr vector<int, Shape.planeDimensions> offset3,
-        constexpr vector<int, Shape.planeDimensions> offset4)
+        constexpr vector<int, Shape.planeDimensions> offset4
+        $(statusDecl))
     {
+        $(statusInit)
         __target_switch
         {
         case hlsl: __intrinsic_asm ".Gather$(cmp)$(componentName)";
@@ -2373,10 +2382,11 @@ ${{{{
             return __texture_gather$(cmp)_offsets<T>(this,$(samplerStateParam) location $(compareArg), offset1,offset2,offset3,offset4 $(glslComponent));
         }
     }
-    ${{{{
+${{{{
+            } // for (isStatus)
         } // for (componentId)
     } // for (isCmp)
-    }}}}
+}}}}
 } // end extension for gather
 
 ${{{{

diff --git a/tests/hlsl-intrinsic/texture/texture-intrinsics.slang b/tests/hlsl-intrinsic/texture/texture-intrinsics.slang
@@ -43,6 +43,8 @@ void computeMain(int3 dispatchThreadID: SV_DispatchThreadID)
     uint numLevels = 0, elements = 0;
     float fnumLevels = 0.0f, felements = 0.0f;
 
+    uint status;
+
     /*
         <Template Type> Object.SampleLevel()
     */     
@@ -244,6 +246,26 @@ void computeMain(int3 dispatchThreadID: SV_DispatchThreadID)
     f4 = t2DArray.Gather(samplerState, float3(u, u, 0), int2(0, 0));
     val += f4.x;  val += f4.y;  val += f4.z;  val += f4.w;
 
+    // Object.GatherGreen()
+    f4 = t2D.GatherGreen(samplerState, float2(u, u), int2(1, 2));
+    val += f4.x;  val += f4.y;  val += f4.z;  val += f4.w;
+
+    f4 = t2D.GatherGreen(samplerState, float2(u, u), int2(1, 2), int2(3, 4), int2(5, 6), int2(7, 8));
+    val += f4.x;  val += f4.y;  val += f4.z;  val += f4.w;
+
+    f4 = tCubeArray.GatherGreen(samplerState, float4(1.5, 1.5, 1.5, 1.5));
+    val += f4.x;  val += f4.y;  val += f4.z;  val += f4.w;
+
+    // status variant
+    f4 = t2D.GatherGreen(samplerState, float2(u, u), int2(1, 2), status);
+    val += f4.x;  val += f4.y;  val += f4.z;  val += f4.w;
+
+    f4 = t2D.GatherGreen(samplerState, float2(u, u), int2(1, 2), int2(3, 4), int2(5, 6), int2(7, 8), status);
+    val += f4.x;  val += f4.y;  val += f4.z;  val += f4.w;
+
+    f4 = tCubeArray.GatherGreen(samplerState, float4(1.5, 1.5, 1.5, 1.5), status);
+    val += f4.x;  val += f4.y;  val += f4.z;  val += f4.w;
+
     /*
         ret Object.Load()
     */
@@ -289,15 +311,15 @@ void computeMain(int3 dispatchThreadID: SV_DispatchThreadID)
 // DX11: 313
 // DX11: 313
 // DX11: 313
-// DX12: 316
-// DX12: 316
-// DX12: 316
-// DX12: 316
-// DX12CS6: 339
-// DX12CS6: 339
-// DX12CS6: 339
-// DX12CS6: 339
-// VK: 339
-// VK: 339
-// VK: 339
-// VK: 339
+// DX12: 340
+// DX12: 340
+// DX12: 340
+// DX12: 340
+// DX12CS6: 363
+// DX12CS6: 363
+// DX12CS6: 363
+// DX12CS6: 363
+// VK: 363
+// VK: 363
+// VK: 363
+// VK: 363