diff --git a/.github/workflows/test_qemu.yml b/.github/workflows/test_qemu.yml
index c2bffbd..cb97092 100644
--- a/.github/workflows/test_qemu.yml
+++ b/.github/workflows/test_qemu.yml
@@ -30,7 +30,7 @@ jobs:
       uses: actions/checkout@v4
       
     - name: Test
-      run: go test -v -short ./...
+      run: go test -v -short ./sm3/...
       env:
         DISABLE_SM3NI: 1
         DISABLE_SM4NI: 1
diff --git a/sm3/sm3blocks_arm64.s b/sm3/sm3blocks_arm64.s
index 61805e3..89bdbc9 100644
--- a/sm3/sm3blocks_arm64.s
+++ b/sm3/sm3blocks_arm64.s
@@ -112,11 +112,11 @@
 	VST1.P [V11.S4], 16(wordPtr)     \
 
 #define ROUND_12_15(index, const, a, b, c, d, e, f, g, h) \
-	MESSAGE_SCHEDULE(index)                        \
+	MESSAGE_SCHEDULE(index)                               \
 	ROUND_00_11(index, const, a, b, c, d, e, f, g, h)     \
 
 #define ROUND_16_63(index, const, a, b, c, d, e, f, g, h) \
-	MESSAGE_SCHEDULE(index); \ // V11 is Wt+4 now, Pls do not use it
+	MESSAGE_SCHEDULE(index)          \ // V11 is Wt+4 now, Pls do not use it
 	PROLD(a, V12, 12)                \
 	VMOV V12.B16, V13.B16            \
 	LOAD_T(const, tmp1)              \
diff --git a/sm3/sm3blocks_test.go b/sm3/sm3blocks_test.go
index 46fbe19..2311304 100644
--- a/sm3/sm3blocks_test.go
+++ b/sm3/sm3blocks_test.go
@@ -3,6 +3,7 @@
 package sm3
 
 import (
+	"encoding/binary"
 	"fmt"
 	"testing"
 )
@@ -53,6 +54,13 @@ func TestBlockMultBy4(t *testing.T) {
 	buffer := make([]byte, 1216)
 	blockMultBy4(&digs[0], &p[0], &buffer[0], 1)
 	expected := "[66c7f0f4 62eeedd9 d1f2d46b dc10e4e2 4167c487 5cf2f7a2 297da02b 8f4ba8e0]"
+	for i := 128; i < 128+68*4*4; i += 64 {
+		fmt.Printf("%08x %08x %08x %08x ", binary.LittleEndian.Uint32(buffer[i:]), binary.LittleEndian.Uint32(buffer[i+4:]), binary.LittleEndian.Uint32(buffer[i+8:]), binary.LittleEndian.Uint32(buffer[i+12:]))
+		fmt.Printf("%08x %08x %08x %08x ", binary.LittleEndian.Uint32(buffer[i+16:]), binary.LittleEndian.Uint32(buffer[i+20:]), binary.LittleEndian.Uint32(buffer[i+24:]), binary.LittleEndian.Uint32(buffer[i+28:]))
+		fmt.Printf("%08x %08x %08x %08x ", binary.LittleEndian.Uint32(buffer[i+32:]), binary.LittleEndian.Uint32(buffer[i+36:]), binary.LittleEndian.Uint32(buffer[i+40:]), binary.LittleEndian.Uint32(buffer[i+44:]))
+		fmt.Printf("%08x %08x %08x %08x ", binary.LittleEndian.Uint32(buffer[i+48:]), binary.LittleEndian.Uint32(buffer[i+52:]), binary.LittleEndian.Uint32(buffer[i+56:]), binary.LittleEndian.Uint32(buffer[i+60:]))
+		fmt.Println()
+	}
 	s := fmt.Sprintf("%x", digs[0][:])
 	if s != expected {
 		t.Errorf("digs[0] got %s", s)