shader: support V_CMPX_NGT_F16 SDWA

src/graphics/shader/recompiler/frontend/decode/ShaderDecoder.h

@@ -366,6 +366,7 @@ enum class Opcode {
 	V_CMPX_LE_F16,
 	V_CMPX_GT_F16,
 	V_CMPX_GE_F16,
+	V_CMPX_NGT_F16,
 	V_CMPX_NEQ_F16,
 	V_CMPX_NLT_F16,
 	V_CMPX_LT_I32,

src/graphics/shader/recompiler/frontend/decode/VectorAluOps.cpp

@@ -242,7 +242,8 @@ constexpr VopcOpcodeInfo VOPC_OPCODE_LIST[] = {
     {0xedu, Opcode::V_CMP_NEQ_F16},       {0xd9u, Opcode::V_CMPX_LT_F16},
     {0xdau, Opcode::V_CMPX_EQ_F16},       {0xdbu, Opcode::V_CMPX_LE_F16},
     {0xdcu, Opcode::V_CMPX_GT_F16},       {0xdeu, Opcode::V_CMPX_GE_F16},
-    {0xfdu, Opcode::V_CMPX_NEQ_F16},      {0xfeu, Opcode::V_CMPX_NLT_F16},
+    {0xfbu, Opcode::V_CMPX_NGT_F16},      {0xfdu, Opcode::V_CMPX_NEQ_F16},
+    {0xfeu, Opcode::V_CMPX_NLT_F16},
 };
 
 constexpr auto VOPC_OPS = Detail::MakeOpcodeTable<0x100>(VOPC_OPCODE_LIST);
@@ -874,6 +875,7 @@ bool IsVopcFloatCompareOpcode(Opcode opcode) {
 		case Opcode::V_CMPX_LE_F16:
 		case Opcode::V_CMPX_GT_F16:
 		case Opcode::V_CMPX_GE_F16:
+		case Opcode::V_CMPX_NGT_F16:
 		case Opcode::V_CMPX_NEQ_F16:
 		case Opcode::V_CMPX_NLT_F16:
 		case Opcode::V_CMP_CLASS_F32: return true;
@@ -1513,6 +1515,7 @@ bool IsVopcCompareExec(Opcode opcode) {
 		case Opcode::V_CMPX_LE_F16:
 		case Opcode::V_CMPX_GT_F16:
 		case Opcode::V_CMPX_GE_F16:
+		case Opcode::V_CMPX_NGT_F16:
 		case Opcode::V_CMPX_NEQ_F16:
 		case Opcode::V_CMPX_NLT_F16: return true;
 		default: return false;

src/graphics/shader/recompiler/frontend/translate/Vector.cpp

@@ -256,6 +256,9 @@ bool Translator::EmitVector(const Decoder::Instruction& inst, std::string* error
 		case O::V_CMPX_LE_F16:
 			EmitFloatCompare(inst, IR::ValueOpcode::FPOrdLessThanEqual32, true, true);
 			return true;
+		case O::V_CMPX_NGT_F16:
+			EmitFloatCompare(inst, IR::ValueOpcode::FPUnordLessThanEqual32, true, true);
+			return true;
 		case O::V_CMP_NEQ_F16:
 			EmitFloatCompare(inst, IR::ValueOpcode::FPUnordNotEqual32, true, false);
 			return true;

tests/ShaderRecompilerComputeTests.cpp

@@ -15772,6 +15772,32 @@ TestCase VectorVopcCmpxGtU16CapturedSdwaExecMask() {
   return test;
 }
 
+TestCase VectorVopcCmpxNgtF16CapturedSdwaExecMask() {
+  using O = ShaderOpcode;
+
+  std::vector<u32> code;
+  AppendVMovLiteral(&code, 31, 0x7e003c00u);
+  AppendVMovU32(&code, 3, 7);
+  AppendVMovU32(&code, 30, 0);
+  code.insert(code.end(), {0x7df700f9u, 0x8605001fu});
+  AppendBufferStoreDword(&code, 3, 30);
+  AppendVMovLiteral(&code, 31, 0x3c007e00u);
+  AppendVMovU32(&code, 3, 9);
+  AppendVMovU32(&code, 30, 4);
+  code.insert(code.end(), {0x7df700f9u, 0x8605001fu});
+  AppendBufferStoreDword(&code, 3, 30);
+  AppendEnd(&code);
+
+  TestCase test{"VectorVopcCmpxNgtF16CapturedSdwaExecMask",
+                code,
+                {0x11111111u, 0x22222222u},
+                {7, 0x22222222u},
+                {O::V_MOV_B32, O::V_CMPX_NGT_F16, O::BUFFER_STORE_DWORD,
+                 O::S_ENDPGM}};
+  test.decoded_counts = {{"V_CMPX_NGT_F16", 2}};
+  return test;
+}
+
 TestCase VectorCompareInvertedMaskSelect() {
   using O = ShaderOpcode;
 
@@ -20297,6 +20323,7 @@ std::vector<TestCase> MakeCases() {
   AddCase(VectorVop3CmpxWritesExecMask);
   AddCase(VectorVopcSdwaCmpxWritesExecMask);
   AddCase(VectorVopcCmpxGtU16CapturedSdwaExecMask);
+  AddCase(VectorVopcCmpxNgtF16CapturedSdwaExecMask);
   AddCase(VectorCompareInvertedMaskSelect);
   AddCase(BranchSelect);
   AddCase(SimpleLoop);