diff --git a/MFMM.md b/MFMM.md
index e3450e8..4f3f99f 100644
--- a/MFMM.md
+++ b/MFMM.md
@@ -116,13 +116,17 @@ acc0, acc1, acc2, acc3, acc4, acc5是64位寄存器
 	MUL	const3, acc0, t0    // t0 = L(acc0*p3)
 	ADCS	t0, acc3, acc3      // (carry3,acc3) = acc3 + L(acc0*p3)
 
-	UMULH	const3, acc0, hlp1 // hlp1 = H(acc0*p3)
+	UMULH	const3, acc0, hlp1 // hlp1 = H(acc0*p3), 事实上不能用hlp1, 这个寄存器被p256PointAddAsm方法全局使用
 	ADC	$0, acc4            // acc4 = carry3 + acc4
 
 	ADDS	acc0, acc1, acc1  // (carry4, acc1) = acc0 + acc1 + L(acc0*p1)
 	ADCS	y0, acc2, acc2    // (carry5, acc2) = carry4 + acc2 +  L(acc0*p2) + H(acc0*p1)
 	ADCS	hlp0, acc3, acc3  // (carry6, acc3) = carry5 + acc3 + L(acc0*p3) + H(acc0*p2)
 	ADC	$0, hlp1, acc0    // acc0 = carry6 + H(acc0*p3)
+
+手上没有arm64环境，只能依赖Travis CI检验代码，但是很慢，效率很低。
+
+
     ======
     用加减替代乘法，但存在潜在风险，进位/借位处理太复杂，所以该实现已经被回滚
     p*acc0 = acc0*2^256 -(acc0*2^32)*2^192 + (acc0 - acc0*2^32)*2^64 - acc0