Complete gfx115X and gfx120X Tensile support

shared/tensile/Tensile/AsmCaps.py

@@ -972,6 +972,7 @@ def nested_update_(left, right):
              (11, 0, 1): {'VOP3v_dot4_i32_i8': True},
              (11, 0, 2): {'VOP3v_dot4_i32_i8': True},
              (11, 0, 3): {'VOP3v_dot4_i32_i8': True},
+             (11, 5, 0): {'VOP3v_dot4_i32_i8': True},
              (11, 5, 1): {'VOP3v_dot4_i32_i8': True},
              (12, 0, 0): {'VOP3v_dot4_i32_i8': True},
              (12, 0, 1): {'VOP3v_dot4_i32_i8': True}}

shared/tensile/Tensile/Common.py

@@ -2464,7 +2464,7 @@ def assignGlobalParameters( config, capabilitiesCache: Optional[dict] = None ):
     if os.name == "nt":
       globalParameters["CurrentISA"] = (9,0,6)
       printWarning("Failed to detect ISA so forcing (gfx906) on windows")
-  isasWithDisabledHWMonitor = ((9,4,2), (9,5,0), (11,0,0), (11,0,1), (11,0,2), (11,0,3), (12,0,0), (12,0,1))
+  isasWithDisabledHWMonitor = ((9,4,2), (9,5,0), (11,0,0), (11,0,1), (11,0,2), (11,0,3), (11,5,0), (11,5,1), (12,0,0), (12,0,1))
   if globalParameters["CurrentISA"] in isasWithDisabledHWMonitor:
     isaString = ', '.join(map(gfxName, isasWithDisabledHWMonitor))
     printWarning(f"HardwareMonitor currently disabled for {isaString}")

shared/tensile/Tensile/Source/lib/include/Tensile/AMDGPU.hpp

@@ -75,7 +75,9 @@ namespace Tensile
             gfx1102 = 1102,
             gfx1103 = 1103,
             gfx1150 = 1150,
-            gfx1151 = 1151
+            gfx1151 = 1151,
+            gfx1200 = 1200,
+            gfx1201 = 1201
         };
 
         static std::string toString(Processor p)
@@ -124,6 +126,10 @@ namespace Tensile
                 return "gfx1150";
             case AMDGPU::Processor::gfx1151:
                 return "gfx1151";
+            case AMDGPU::Processor::gfx1200:
+                return "gfx1200";
+            case AMDGPU::Processor::gfx1201:
+                return "gfx1201";
             }
             return "";
         }
@@ -198,6 +204,14 @@ namespace Tensile
             {
                 return AMDGPU::Processor::gfx1151;
             }
+            else if(deviceString.find("gfx1200") != std::string::npos)
+            {
+                return AMDGPU::Processor::gfx1200;
+            }
+            else if(deviceString.find("gfx1201") != std::string::npos)
+            {
+                return AMDGPU::Processor::gfx1201;
+            }
             else
             {
                 return static_cast<AMDGPU::Processor>(0);

shared/tensile/Tensile/Source/lib/include/Tensile/PlaceholderLibrary.hpp

@@ -60,6 +60,8 @@ namespace Tensile
         gfx1103,
         gfx1150,
         gfx1151,
+        gfx1200,
+        gfx1201,
         All
     };
 
@@ -112,6 +114,10 @@ namespace Tensile
             return "TensileLibrary_*_gfx1150";
         case LazyLoadingInit::gfx1151:
             return "TensileLibrary_*_gfx1151";
+        case LazyLoadingInit::gfx1200:
+            return "TensileLibrary_*_gfx1200";
+        case LazyLoadingInit::gfx1201:
+            return "TensileLibrary_*_gfx1201";
         case LazyLoadingInit::None:
             return "";
         }

shared/tensile/Tensile/Source/lib/include/Tensile/Serialization/Predicates.hpp

@@ -236,6 +236,8 @@ namespace Tensile
                 iot::enumCase(io, value, "gfx1103", AMDGPU::Processor::gfx1103);
                 iot::enumCase(io, value, "gfx1150", AMDGPU::Processor::gfx1150);
                 iot::enumCase(io, value, "gfx1151", AMDGPU::Processor::gfx1151);
+                iot::enumCase(io, value, "gfx1200", AMDGPU::Processor::gfx1200);
+                iot::enumCase(io, value, "gfx1201", AMDGPU::Processor::gfx1201);
             }
         };
 

shared/tensile/docs/src/cli-reference/tensile-create-library-cli.rst

@@ -56,7 +56,7 @@ Here is the list of optional arguments for invoking the ``TensileCreateLibrary``
       Supported architectures include: all; gfx000; gfx803; gfx900; gfx900:xnack-; gfx906; gfx906:xnack+; gfx906:xnack-; gfx908; gfx908:xnack+;
       gfx908:xnack-; gfx90a; gfx90a:xnack+; gfx90a:xnack-; gfx940; gfx940:xnack+; gfx940:xnack-; gfx941; gfx941:xnack+;
       gfx941:xnack-; gfx942; gfx942:xnack+; gfx942:xnack-; gfx1010; gfx1011; gfx1012; gfx1030; gfx1031; gfx1032; gfx1034; gfx1035;
-      gfx1100; gfx1101; gfx1102; gfx1103.
+      gfx1100; gfx1101; gfx1102; gfx1103; gfx1150; gfx1151; gfx1200; gfx1201.
 
   * - \-\-build-client
     - Builds Tensile client executable that is used for stand alone benchmarking. This option is set by default.