diff --git a/src/backend_x64/emit_x64_vector.cpp b/src/backend_x64/emit_x64_vector.cpp
index 206e3b74..23c17d23 100644
--- a/src/backend_x64/emit_x64_vector.cpp
+++ b/src/backend_x64/emit_x64_vector.cpp
@@ -5,6 +5,7 @@
  */
 
 #include <algorithm>
+#include <bitset>
 #include <functional>
 #include <type_traits>
 
@@ -1865,6 +1866,27 @@ void EmitX64::EmitVectorPairedAddUnsignedWiden32(EmitContext& ctx, IR::Inst* ins
     ctx.reg_alloc.DefineValue(inst, a);
 }
 
+template <typename T>
+static T PolynomialMultiply(T lhs, T rhs) {
+    constexpr size_t bit_size = Common::BitSize<T>();
+    const std::bitset<bit_size> operand(lhs);
+
+    T res = 0;
+    for (size_t i = 0; i < bit_size; i++) {
+        if (operand[i]) {
+            res ^= rhs << i;
+        }
+    }
+
+    return res;
+}
+
+void EmitX64::EmitVectorPolynomialMultiply8(EmitContext& ctx, IR::Inst* inst) {
+    EmitTwoArgumentFallback(code, ctx, inst, [](VectorArray<u8>& result, const VectorArray<u8>& a, const VectorArray<u8>& b) {
+        std::transform(a.begin(), a.end(), b.begin(), result.begin(), PolynomialMultiply<u8>);
+    });
+}
+
 void EmitX64::EmitVectorPopulationCount(EmitContext& ctx, IR::Inst* inst) {
     if (code.DoesCpuSupport(Xbyak::util::Cpu::tAVX512_BITALG)) {
         auto args = ctx.reg_alloc.GetArgumentInfo(inst);
diff --git a/src/frontend/ir/ir_emitter.cpp b/src/frontend/ir/ir_emitter.cpp
index b3001b5f..120775a6 100644
--- a/src/frontend/ir/ir_emitter.cpp
+++ b/src/frontend/ir/ir_emitter.cpp
@@ -1194,6 +1194,10 @@ U128 IREmitter::VectorPairedAddUnsignedWiden(size_t original_esize, const U128&
     return {};
 }
 
+U128 IREmitter::VectorPolynomialMultiply(const U128& a, const U128& b) {
+    return Inst<U128>(Opcode::VectorPolynomialMultiply8, a, b);
+}
+
 U128 IREmitter::VectorPopulationCount(const U128& a) {
     return Inst<U128>(Opcode::VectorPopulationCount, a);
 }
diff --git a/src/frontend/ir/ir_emitter.h b/src/frontend/ir/ir_emitter.h
index 25bdaff1..5e66712a 100644
--- a/src/frontend/ir/ir_emitter.h
+++ b/src/frontend/ir/ir_emitter.h
@@ -238,6 +238,7 @@ public:
     U128 VectorPairedAddLower(size_t esize, const U128& a, const U128& b);
     U128 VectorPairedAddSignedWiden(size_t original_esize, const U128& a);
     U128 VectorPairedAddUnsignedWiden(size_t original_esize, const U128& a);
+    U128 VectorPolynomialMultiply(const U128& a, const U128& b);
     U128 VectorPopulationCount(const U128& a);
     U128 VectorReverseBits(const U128& a);
     U128 VectorRotateLeft(size_t esize, const U128& a, u8 amount);
diff --git a/src/frontend/ir/opcodes.inc b/src/frontend/ir/opcodes.inc
index 9ab17dcd..2e2534b2 100644
--- a/src/frontend/ir/opcodes.inc
+++ b/src/frontend/ir/opcodes.inc
@@ -330,6 +330,7 @@ OPCODE(VectorPairedAdd8,                        T::U128,        T::U128,
 OPCODE(VectorPairedAdd16,                       T::U128,        T::U128,        T::U128                         )
 OPCODE(VectorPairedAdd32,                       T::U128,        T::U128,        T::U128                         )
 OPCODE(VectorPairedAdd64,                       T::U128,        T::U128,        T::U128                         )
+OPCODE(VectorPolynomialMultiply8,               T::U128,        T::U128,        T::U128                         )
 OPCODE(VectorPopulationCount,                   T::U128,        T::U128                                         )
 OPCODE(VectorReverseBits,                       T::U128,        T::U128                                         )
 OPCODE(VectorRoundingHalvingAddS8,              T::U128,        T::U128,        T::U128                         )