diff --git a/sm4/asm_arm64.s b/sm4/asm_arm64.s
index 0ec3f62..62fd329 100644
--- a/sm4/asm_arm64.s
+++ b/sm4/asm_arm64.s
@@ -226,85 +226,6 @@ TEXT ·encryptBlocksAsm(SB),NOSPLIT,$0
   VEOR ZERO.B16, ZERO.B16, ZERO.B16
   
   EOR R0, R0
-
-encryptBlocksLoop:
-  MOVW.P 4(R8), R19
-  VMOV R19, x.S[0]
-  VMOV R19, x.S[1]
-  VMOV R19, x.S[2]
-  VMOV R19, x.S[3]
-  VEOR t1.B16, x.B16, x.B16
-  VEOR t2.B16, x.B16, x.B16
-  VEOR t3.B16, x.B16, x.B16
-  SM4_TAO_L1(x, y)
-  VEOR x.B16, t0.B16, t0.B16
-
-  MOVW.P 4(R8), R19
-  VMOV R19, x.S[0]
-  VMOV R19, x.S[1]
-  VMOV R19, x.S[2]
-  VMOV R19, x.S[3]
-  VEOR t0.B16, x.B16, x.B16
-  VEOR t2.B16, x.B16, x.B16
-  VEOR t3.B16, x.B16, x.B16
-  SM4_TAO_L1(x, y)
-  VEOR x.B16, t1.B16, t1.B16
-
-  MOVW.P 4(R8), R19
-  VMOV R19, x.S[0]
-  VMOV R19, x.S[1]
-  VMOV R19, x.S[2]
-  VMOV R19, x.S[3]
-  VEOR t0.B16, x.B16, x.B16
-  VEOR t1.B16, x.B16, x.B16
-  VEOR t3.B16, x.B16, x.B16
-  SM4_TAO_L1(x, y)
-  VEOR x.B16, t2.B16, t2.B16
-  ADD $4, R0
-
-  MOVW.P 4(R8), R19
-  VMOV R19, x.S[0]
-  VMOV R19, x.S[1]
-  VMOV R19, x.S[2]
-  VMOV R19, x.S[3]
-  VEOR t0.B16, x.B16, x.B16
-  VEOR t1.B16, x.B16, x.B16
-  VEOR t2.B16, x.B16, x.B16
-  SM4_TAO_L1(x, y)
-  VEOR x.B16, t3.B16, t3.B16
-
-  ADD $16, R0
-  CMP $128, R0
-  BNE encryptBlocksLoop
-
-  VTBL FLIP_MASK.B16, [t0.B16], t0.B16
-  VTBL FLIP_MASK.B16, [t1.B16], t1.B16
-  VTBL FLIP_MASK.B16, [t2.B16], t2.B16
-  VTBL FLIP_MASK.B16, [t3.B16], t3.B16
-
-  VMOV t3.S[0], V8.S[0]
-  VMOV t2.S[0], V8.S[1]
-  VMOV t1.S[0], V8.S[2]
-  VMOV t0.S[0], V8.S[3]
-  VST1.P	[V8.B16], 16(R9)
-
-  VMOV t3.S[1], V8.S[0]
-  VMOV t2.S[1], V8.S[1]
-  VMOV t1.S[1], V8.S[2]
-  VMOV t0.S[1], V8.S[3]
-  VST1.P	[V8.B16], 16(R9)
-
-  VMOV t3.S[2], V8.S[0]
-  VMOV t2.S[2], V8.S[1]
-  VMOV t1.S[2], V8.S[2]
-  VMOV t0.S[2], V8.S[3]
-  VST1.P	[V8.B16], 16(R9)
-
-  VMOV t3.S[3], V8.S[0]
-  VMOV t2.S[3], V8.S[1]
-  VMOV t1.S[3], V8.S[2]
-  VMOV t0.S[3], V8.S[3]
-  VST1	[V8.B16], (R9)
 	RET
 
 
@@ -330,64 +251,4 @@ TEXT ·encryptBlockAsm(SB),NOSPLIT,$0
 	VMOV	R0, FLIP_MASK.D[0]
 	VMOV	R1, FLIP_MASK.D[1]
   EOR R0, R0
-
-encryptBlockLoop:
-  MOVW.P 4(R8), R19
-  VMOV R19, x.S[0]
-  VMOV R19, x.S[1]
-  VMOV R19, x.S[2]
-  VMOV R19, x.S[3]
-  VEOR t1.B16, x.B16, x.B16
-  VEOR t2.B16, x.B16, x.B16
-  VEOR t3.B16, x.B16, x.B16
-  SM4_TAO_L1(x, y)
-  VEOR x.B16, t0.B16, t0.B16
-
-  MOVW.P 4(R8), R19
-  VMOV R19, x.S[0]
-  VMOV R19, x.S[1]
-  VMOV R19, x.S[2]
-  VMOV R19, x.S[3]
-  VEOR t0.B16, x.B16, x.B16
-  VEOR t2.B16, x.B16, x.B16
-  VEOR t3.B16, x.B16, x.B16
-  SM4_TAO_L1(x, y)
-  VEOR x.B16, t1.B16, t1.B16
-
-  MOVW.P 4(R8), R19
-  VMOV R19, x.S[0]
-  VMOV R19, x.S[1]
-  VMOV R19, x.S[2]
-  VMOV R19, x.S[3]
-  VEOR t0.B16, x.B16, x.B16
-  VEOR t1.B16, x.B16, x.B16
-  VEOR t3.B16, x.B16, x.B16
-  SM4_TAO_L1(x, y)
-  VEOR x.B16, t2.B16, t2.B16
-
-  MOVW.P 4(R8), R19
-  VMOV R19, x.S[0]
-  VMOV R19, x.S[1]
-  VMOV R19, x.S[2]
-  VMOV R19, x.S[3]
-  VEOR t0.B16, x.B16, x.B16
-  VEOR t1.B16, x.B16, x.B16
-  VEOR t2.B16, x.B16, x.B16
-  SM4_TAO_L1(x, y)
-  VEOR x.B16, t3.B16, t3.B16
-
-  ADD $16, R0
-  CMP $128, R0
-  BNE encryptBlockLoop
-
-  VTBL FLIP_MASK.B16, [t0.B16], t0.B16
-  VTBL FLIP_MASK.B16, [t1.B16], t1.B16
-  VTBL FLIP_MASK.B16, [t2.B16], t2.B16
-  VTBL FLIP_MASK.B16, [t3.B16], t3.B16
-
-  VMOV t3.S[0], V8.S[0]
-  VMOV t2.S[0], V8.S[1]
-  VMOV t1.S[0], V8.S[2]
-  VMOV t0.S[0], V8.S[3]
-  VST1	[V8.B16], (R9)
 	RET