diff --git a/kernel/loongarch64/KERNEL.LOONGSON2K1000 b/kernel/loongarch64/KERNEL.LOONGSON2K1000
index 565bec0f2..879a6f68b 100644
--- a/kernel/loongarch64/KERNEL.LOONGSON2K1000
+++ b/kernel/loongarch64/KERNEL.LOONGSON2K1000
@@ -34,4 +34,7 @@ IDAMINKERNEL = idamin_lsx.S
 SCOPYKERNEL =  scopy_lsx.S
 DCOPYKERNEL =  dcopy_lsx.S
 
+SSWAPKERNEL =  sswap_lsx.S
+DSWAPKERNEL =  dswap_lsx.S
+
 endif
diff --git a/kernel/loongarch64/KERNEL.LOONGSON3R5 b/kernel/loongarch64/KERNEL.LOONGSON3R5
index a2443720b..581cfdbbe 100644
--- a/kernel/loongarch64/KERNEL.LOONGSON3R5
+++ b/kernel/loongarch64/KERNEL.LOONGSON3R5
@@ -34,6 +34,9 @@ IDAMINKERNEL = idamin_lasx.S
 SCOPYKERNEL =  scopy_lasx.S
 DCOPYKERNEL =  dcopy_lasx.S
 
+SSWAPKERNEL =  sswap_lasx.S
+DSWAPKERNEL =  dswap_lasx.S
+
 DGEMMKERNEL    = dgemm_kernel_16x4.S
 DGEMMINCOPY    = dgemm_ncopy_16.S
 DGEMMITCOPY    = dgemm_tcopy_16.S
diff --git a/kernel/loongarch64/dswap_lasx.S b/kernel/loongarch64/dswap_lasx.S
new file mode 100644
index 000000000..221cb7fa2
--- /dev/null
+++ b/kernel/loongarch64/dswap_lasx.S
@@ -0,0 +1,301 @@
+#define ASSEMBLER
+
+#include "common.h"
+#define N      $r4
+#define X      $r7
+#define INCX   $r8
+#define Y      $r9
+#define INCY   $r10
+
+#define I      $r17
+#define TEMP   $r18
+#define XX     $r5
+#define YY     $r6
+#define t1     $r14
+#define t2     $r15
+#define t3     $r16
+#define t4     $r19
+#define a1     $f12
+#define a2     $f13
+#define a3     $f14
+#define a4     $f15
+#define b1     $f16
+#define b2     $f17
+#define b3     $f18
+#define b4     $f19
+#define VX0    $xr12
+#define VX1    $xr13
+#define VX2    $xr14
+#define VX3    $xr15
+
+
+    PROLOGUE
+    bge $r0, N, .L999
+    li.d TEMP, 1
+    slli.d  TEMP, TEMP, BASE_SHIFT
+    slli.d  INCX, INCX, BASE_SHIFT
+    slli.d  INCY, INCY, BASE_SHIFT
+    srai.d I, N, 3
+    bne INCX, TEMP, .L20
+    bne INCY, TEMP, .L12 // INCX==1 and INCY!=1
+    b .L11  // INCX==1 and INCY==1
+.L20:
+    bne INCY, TEMP, .L22 // INCX!=1 and INCY!=1
+    b .L21 // INCX!=1 and INCY==1
+
+.L11:
+    bge $r0, I, .L112
+    .align 3
+
+.L111:
+    xvld VX0, X, 0 * SIZE
+    xvld VX1, X, 4 * SIZE
+    xvld VX2, Y, 0 * SIZE
+    xvld VX3, Y, 4 * SIZE
+    addi.d  I, I, -1
+    xvst VX2, X, 0 * SIZE
+    xvst VX3, X, 4 * SIZE
+    xvst VX0, Y, 0 * SIZE
+    xvst VX1, Y, 4 * SIZE
+    addi.d X, X, 8 * SIZE
+    addi.d Y, Y, 8 * SIZE
+    blt $r0, I, .L111
+    .align 3
+
+.L112:
+    andi I, N, 7
+    bge $r0, I, .L999
+    .align 3
+
+.L113:
+    fld.d $f12, X, 0 * SIZE
+    fld.d $f14, Y, 0 * SIZE
+    addi.d I, I, -1
+    fst.d $f12, Y, 0 * SIZE
+    fst.d $f14, X, 0 * SIZE
+    addi.d  X, X, SIZE
+    addi.d  Y, Y, SIZE
+    blt $r0, I, .L113
+    b .L999
+    .align 3
+
+.L12: // INCX==1 and INCY!=1
+    bge $r0, I, .L122
+    .align 3
+
+.L121:
+    xvld VX0, X, 0 * SIZE
+    ld.d t1, Y, 0 * SIZE
+    xvstelm.d VX0, Y, 0, 0
+    add.d Y, Y, INCY
+    ld.d t2, Y, 0 * SIZE
+    xvstelm.d VX0, Y, 0, 1
+    add.d Y, Y, INCY
+    ld.d t3, Y, 0 * SIZE
+    xvstelm.d VX0, Y, 0, 2
+    add.d Y, Y, INCY
+    ld.d t4, Y, 0 * SIZE
+    xvstelm.d VX0, Y, 0, 3
+    xvinsgr2vr.d VX2, t1, 0
+    xvinsgr2vr.d VX2, t2, 1
+    xvinsgr2vr.d VX2, t3, 2
+    xvinsgr2vr.d VX2, t4, 3
+    add.d Y, Y, INCY
+    xvst VX2, X, 0 * SIZE
+    xvld VX1, X, 4 * SIZE
+    ld.d t1, Y, 0 * SIZE
+    xvstelm.d VX1, Y, 0, 0
+    add.d Y, Y, INCY
+    ld.d t2, Y, 0 * SIZE
+    xvstelm.d VX1, Y, 0, 1
+    add.d Y, Y, INCY
+    ld.d t3, Y, 0 * SIZE
+    xvstelm.d VX1, Y, 0, 2
+    add.d Y, Y, INCY
+    ld.d t4, Y, 0 * SIZE
+    xvstelm.d VX1, Y, 0, 3
+    xvinsgr2vr.d VX3, t1, 0
+    xvinsgr2vr.d VX3, t2, 1
+    xvinsgr2vr.d VX3, t3, 2
+    xvinsgr2vr.d VX3, t4, 3
+    add.d Y, Y, INCY
+    xvst VX3, X, 4 * SIZE
+    addi.d X, X, 8 * SIZE
+    addi.d  I, I, -1
+    blt $r0, I, .L121
+    .align 3
+
+.L122:
+    andi I, N, 7
+    bge $r0, I, .L999
+    .align 3
+
+.L123:
+    fld.d $f12, X, 0 * SIZE
+    fld.d $f14, Y, 0 * SIZE
+    addi.d I, I, -1
+    fst.d $f12, Y, 0 * SIZE
+    fst.d $f14, X, 0 * SIZE
+    addi.d  X, X, SIZE
+    add.d  Y, Y, INCY
+    blt $r0, I, .L123
+    b .L999
+    .align 3
+
+.L21:
+    bge $r0, I, .L212
+    .align 3
+
+.L211:
+    xvld VX2, Y, 0 * SIZE
+    ld.d t1, X, 0 * SIZE
+    xvstelm.d VX2, X, 0, 0
+    add.d X, X, INCX
+    ld.d t2, X, 0 * SIZE
+    xvstelm.d VX2, X, 0, 1
+    add.d X, X, INCX
+    ld.d t3, X, 0 * SIZE
+    xvstelm.d VX2, X, 0, 2
+    add.d X, X, INCX
+    ld.d t4, X, 0 * SIZE
+    xvstelm.d VX2, X, 0, 3
+    xvinsgr2vr.d VX0, t1, 0
+    xvinsgr2vr.d VX0, t2, 1
+    xvinsgr2vr.d VX0, t3, 2
+    xvinsgr2vr.d VX0, t4, 3
+    add.d X, X, INCX
+    xvst VX0, Y, 0 * SIZE
+    xvld VX3, Y, 4 * SIZE
+    ld.d t1, X, 0 * SIZE
+    xvstelm.d VX3, X, 0, 0
+    add.d X, X, INCY
+    ld.d t2, X, 0 * SIZE
+    xvstelm.d VX3, X, 0, 1
+    add.d X, X, INCX
+    ld.d t3, X, 0 * SIZE
+    xvstelm.d VX3, X, 0, 2
+    add.d X, X, INCX
+    ld.d t4, X, 0 * SIZE
+    xvstelm.d VX3, X, 0, 3
+    xvinsgr2vr.d VX1, t1, 0
+    xvinsgr2vr.d VX1, t2, 1
+    xvinsgr2vr.d VX1, t3, 2
+    xvinsgr2vr.d VX1, t4, 3
+    add.d X, X, INCX
+    xvst VX1, Y, 0 * SIZE
+    addi.d Y, Y, 8 * SIZE
+    addi.d  I, I, -1
+    blt $r0, I, .L211
+    .align 3
+
+.L212:
+    andi I, N, 7
+    bge $r0, I, .L999
+    .align 3
+
+.L213:
+    fld.d $f12, X, 0 * SIZE
+    fld.d $f14, Y, 0 * SIZE
+    addi.d I, I, -1
+    fst.d $f12, Y, 0 * SIZE
+    fst.d $f14, X, 0 * SIZE
+    add.d  X, X, INCX
+    addi.d  Y, Y, SIZE
+    blt $r0, I, .L213
+    b .L999
+    .align 3
+
+.L22:
+    bgez INCX, .L220
+    //addi.d TEMP, N, -1
+    //mul.d TEMP, TEMP, INCX
+    //sub.d X, X, TEMP
+    .align 3
+
+.L220:
+    bge $r0, I, .L223
+    .align 3
+    move XX, X
+
+.L222:
+    fld.d a1, X, 0 * SIZE
+    add.d X, X, INCX
+    fld.d a2, X, 0 * SIZE
+    add.d X, X, INCX
+    fld.d a3, X, 0 * SIZE
+    add.d X, X, INCX
+    fld.d a4, X, 0 * SIZE
+    add.d X, X, INCX
+    fld.d b1, Y, 0 * SIZE
+    fst.d a1, Y, 0 * SIZE
+    add.d Y, Y, INCY
+    fld.d b2, Y, 0 * SIZE
+    fst.d a2, Y, 0 * SIZE
+    add.d Y, Y, INCY
+    fld.d b3, Y, 0 * SIZE
+    fst.d a3, Y, 0 * SIZE
+    add.d Y, Y, INCY
+    fld.d b4, Y, 0 * SIZE
+    fst.d a4, Y, 0 * SIZE
+    add.d Y, Y, INCY
+    fld.d a1, X, 0 * SIZE
+    add.d X, X, INCX
+    fst.d b1, XX, 0 * SIZE
+    add.d XX, XX, INCX
+    fld.d b1, Y, 0 * SIZE
+    fst.d a1, Y, 0 * SIZE
+    add.d Y, Y, INCY
+    fld.d a2, X, 0 * SIZE
+    add.d X, X, INCX
+    fst.d b2, XX, 0 * SIZE
+    add.d XX, XX, INCX
+    fld.d b2, Y, 0 * SIZE
+    fst.d a2, Y, 0 * SIZE
+    add.d Y, Y, INCY
+    fld.d a3, X, 0 * SIZE
+    add.d X, X, INCX
+    fst.d b3, XX, 0 * SIZE
+    add.d XX, XX, INCX
+    fld.d b3, Y, 0 * SIZE
+    fst.d a3, Y, 0 * SIZE
+    fld.d a4, X, 0 * SIZE
+    add.d X, X, INCX
+    fst.d b4, XX, 0 * SIZE
+    add.d XX, XX, INCX
+    fld.d b4, Y, 0 * SIZE
+    fst.d a4, Y, 0 * SIZE
+    add.d Y, Y, INCY
+    fst.d b1, XX, 0 * SIZE
+    add.d XX, XX, INCX
+    fst.d b2, XX, 0 * SIZE
+    add.d XX, XX, INCX
+    fst.d b3, XX, 0 * SIZE
+    add.d XX, XX, INCX
+    fst.d b4, XX, 0 * SIZE
+    add.d XX, XX, INCX
+    addi.d  I, I, -1
+    blt $r0, I, .L222
+    .align 3
+
+.L223:
+    andi I, N, 7
+    bge $r0, I, .L999
+    .align 3
+
+.L224:
+    fld.d $f12, X, 0 * SIZE
+    fld.d $f14, Y, 0 * SIZE
+    addi.d I, I, -1
+    fst.d $f12, Y, 0 * SIZE
+    fst.d $f14, X, 0 * SIZE
+    add.d  X, X, INCX
+    add.d  Y, Y, INCY
+    blt $r0, I, .L224
+    .align 3
+
+.L999:
+    move $r4, $r12
+    jirl $r0, $r1, 0x0
+    .align 3
+
+    EPILOGUE
diff --git a/kernel/loongarch64/dswap_lsx.S b/kernel/loongarch64/dswap_lsx.S
new file mode 100644
index 000000000..7f7f585e1
--- /dev/null
+++ b/kernel/loongarch64/dswap_lsx.S
@@ -0,0 +1,317 @@
+#define ASSEMBLER
+
+#include "common.h"
+#define N      $r4
+#define X      $r7
+#define INCX   $r8
+#define Y      $r9
+#define INCY   $r10
+
+#define I      $r17
+#define TEMP   $r18
+#define XX     $r5
+#define YY     $r6
+#define t1     $r14
+#define t2     $r15
+#define t3     $r16
+#define t4     $r19
+#define a1     $f12
+#define a2     $f13
+#define a3     $f14
+#define a4     $f15
+#define b1     $f16
+#define b2     $f17
+#define b3     $f18
+#define b4     $f19
+#define VX0    $vr12
+#define VX1    $vr13
+#define VX2    $vr14
+#define VX3    $vr15
+
+
+    PROLOGUE
+    bge $r0, N, .L999
+    li.d TEMP, 1
+    slli.d  TEMP, TEMP, BASE_SHIFT
+    slli.d  INCX, INCX, BASE_SHIFT
+    slli.d  INCY, INCY, BASE_SHIFT
+    srai.d I, N, 3
+    bne INCX, TEMP, .L20
+    bne INCY, TEMP, .L12 // INCX==1 and INCY!=1
+    b .L11  // INCX==1 and INCY==1
+.L20:
+    bne INCY, TEMP, .L22 // INCX!=1 and INCY!=1
+    b .L21 // INCX!=1 and INCY==1
+
+.L11:
+    bge $r0, I, .L112
+    .align 3
+
+.L111:
+    vld VX0, X, 0 * SIZE
+    vld VX1, X, 2 * SIZE
+    vld VX2, Y, 0 * SIZE
+    vld VX3, Y, 2 * SIZE
+    vst VX2, X, 0 * SIZE
+    vst VX3, X, 2 * SIZE
+    vst VX0, Y, 0 * SIZE
+    vst VX1, Y, 2 * SIZE
+    vld VX0, X, 4 * SIZE
+    vld VX1, X, 6 * SIZE
+    vld VX2, Y, 4 * SIZE
+    vld VX3, Y, 6 * SIZE
+    addi.d  I, I, -1
+    vst VX2, X, 4 * SIZE
+    vst VX3, X, 6 * SIZE
+    vst VX0, Y, 4 * SIZE
+    vst VX1, Y, 6 * SIZE
+    addi.d X, X, 8 * SIZE
+    addi.d Y, Y, 8 * SIZE
+    blt $r0, I, .L111
+    .align 3
+
+.L112:
+    andi I, N, 7
+    bge $r0, I, .L999
+    .align 3
+
+.L113:
+    fld.d $f12, X, 0 * SIZE
+    fld.d $f14, Y, 0 * SIZE
+    addi.d I, I, -1
+    fst.d $f12, Y, 0 * SIZE
+    fst.d $f14, X, 0 * SIZE
+    addi.d  X, X, SIZE
+    addi.d  Y, Y, SIZE
+    blt $r0, I, .L113
+    b .L999
+    .align 3
+
+.L12: // INCX==1 and INCY!=1
+    bge $r0, I, .L122
+    .align 3
+
+.L121:
+    vld VX0, X, 0 * SIZE
+    ld.d t1, Y, 0 * SIZE
+    vstelm.d VX0, Y, 0, 0
+    add.d Y, Y, INCY
+    ld.d t2, Y, 0 * SIZE
+    vstelm.d VX0, Y, 0, 1
+    vinsgr2vr.d VX2, t1, 0
+    vinsgr2vr.d VX2, t2, 1
+    add.d Y, Y, INCY
+    vst VX2, X, 0 * SIZE
+    vld VX1, X, 2 * SIZE
+    ld.d t3, Y, 0 * SIZE
+    vstelm.d VX1, Y, 0, 0
+    add.d Y, Y, INCY
+    ld.d t4, Y, 0 * SIZE
+    vstelm.d VX1, Y, 0, 1
+    vinsgr2vr.d VX3, t3, 0
+    vinsgr2vr.d VX3, t4, 1
+    add.d Y, Y, INCY
+    vst VX3, X, 2 * SIZE
+    vld VX0, X, 4 * SIZE
+    ld.d t1, Y, 0 * SIZE
+    vstelm.d VX0, Y, 0, 0
+    add.d Y, Y, INCY
+    ld.d t2, Y, 0 * SIZE
+    vstelm.d VX0, Y, 0, 1
+    vinsgr2vr.d VX2, t1, 0
+    vinsgr2vr.d VX2, t2, 1
+    add.d Y, Y, INCY
+    vst VX2, X, 4 * SIZE
+    vld VX1, X, 6 * SIZE
+    ld.d t3, Y, 0 * SIZE
+    vstelm.d VX1, Y, 0, 0
+    add.d Y, Y, INCY
+    ld.d t4, Y, 0 * SIZE
+    vstelm.d VX1, Y, 0, 1
+    vinsgr2vr.d VX3, t3, 0
+    vinsgr2vr.d VX3, t4, 1
+    add.d Y, Y, INCY
+    vst VX3, X, 6 * SIZE
+    addi.d X, X, 8 * SIZE
+    addi.d  I, I, -1
+    blt $r0, I, .L121
+    .align 3
+
+.L122:
+    andi I, N, 7
+    bge $r0, I, .L999
+    .align 3
+
+.L123:
+    fld.d $f12, X, 0 * SIZE
+    fld.d $f14, Y, 0 * SIZE
+    addi.d I, I, -1
+    fst.d $f12, Y, 0 * SIZE
+    fst.d $f14, X, 0 * SIZE
+    addi.d  X, X, SIZE
+    add.d  Y, Y, INCY
+    blt $r0, I, .L123
+    b .L999
+    .align 3
+
+.L21:
+    bge $r0, I, .L212
+    .align 3
+
+.L211:
+    vld VX2, Y, 0 * SIZE
+    ld.d t1, X, 0 * SIZE
+    vstelm.d VX2, X, 0, 0
+    add.d X, X, INCX
+    ld.d t2, X, 0 * SIZE
+    vstelm.d VX2, X, 0, 1
+    vinsgr2vr.d VX0, t1, 0
+    vinsgr2vr.d VX0, t2, 1
+    add.d X, X, INCY
+    vst VX0, Y, 0 * SIZE
+    vld VX3, Y, 2 * SIZE
+    ld.d t3, X, 0 * SIZE
+    vstelm.d VX3, X, 0, 0
+    add.d X, X, INCX
+    ld.d t4, X, 0 * SIZE
+    vstelm.d VX3, X, 0, 1
+    vinsgr2vr.d VX1, t3, 0
+    vinsgr2vr.d VX1, t4, 1
+    add.d X, X, INCX
+    vst VX1, Y, 2 * SIZE
+    vld VX2, Y, 4 * SIZE
+    ld.d t1, X, 0 * SIZE
+    vstelm.d VX2, X, 0, 0
+    add.d X, X, INCX
+    ld.d t2, X, 0 * SIZE
+    vstelm.d VX2, X, 0, 1
+    vinsgr2vr.d VX0, t1, 0
+    vinsgr2vr.d VX0, t2, 1
+    add.d X, X, INCY
+    vst VX0, Y, 4 * SIZE
+    vld VX3, Y, 6 * SIZE
+    ld.d t3, X, 0 * SIZE
+    vstelm.d VX3, X, 0, 0
+    add.d X, X, INCX
+    ld.d t4, X, 0 * SIZE
+    vstelm.d VX3, X, 0, 1
+    vinsgr2vr.d VX1, t3, 0
+    vinsgr2vr.d VX1, t4, 1
+    add.d X, X, INCX
+    vst VX1, Y, 6 * SIZE
+    addi.d Y, Y, 8 * SIZE
+    addi.d  I, I, -1
+    blt $r0, I, .L211
+    .align 3
+
+.L212:
+    andi I, N, 7
+    bge $r0, I, .L999
+    .align 3
+
+.L213:
+    fld.d $f12, X, 0 * SIZE
+    fld.d $f14, Y, 0 * SIZE
+    addi.d I, I, -1
+    fst.d $f12, Y, 0 * SIZE
+    fst.d $f14, X, 0 * SIZE
+    add.d  X, X, INCX
+    addi.d  Y, Y, SIZE
+    blt $r0, I, .L213
+    b .L999
+    .align 3
+
+.L22:
+    bgez INCX, .L220
+    //addi.d TEMP, N, -1
+    //mul.d TEMP, TEMP, INCX
+    //sub.d X, X, TEMP
+    .align 3
+
+.L220:
+    bge $r0, I, .L223
+    .align 3
+    move XX, X
+
+.L222:
+    fld.d a1, X, 0 * SIZE
+    add.d X, X, INCX
+    fld.d a2, X, 0 * SIZE
+    add.d X, X, INCX
+    fld.d a3, X, 0 * SIZE
+    add.d X, X, INCX
+    fld.d a4, X, 0 * SIZE
+    add.d X, X, INCX
+    fld.d b1, Y, 0 * SIZE
+    fst.d a1, Y, 0 * SIZE
+    add.d Y, Y, INCY
+    fld.d b2, Y, 0 * SIZE
+    fst.d a2, Y, 0 * SIZE
+    add.d Y, Y, INCY
+    fld.d b3, Y, 0 * SIZE
+    fst.d a3, Y, 0 * SIZE
+    add.d Y, Y, INCY
+    fld.d b4, Y, 0 * SIZE
+    fst.d a4, Y, 0 * SIZE
+    add.d Y, Y, INCY
+    fld.d a1, X, 0 * SIZE
+    add.d X, X, INCX
+    fst.d b1, XX, 0 * SIZE
+    add.d XX, XX, INCX
+    fld.d b1, Y, 0 * SIZE
+    fst.d a1, Y, 0 * SIZE
+    add.d Y, Y, INCY
+    fld.d a2, X, 0 * SIZE
+    add.d X, X, INCX
+    fst.d b2, XX, 0 * SIZE
+    add.d XX, XX, INCX
+    fld.d b2, Y, 0 * SIZE
+    fst.d a2, Y, 0 * SIZE
+    add.d Y, Y, INCY
+    fld.d a3, X, 0 * SIZE
+    add.d X, X, INCX
+    fst.d b3, XX, 0 * SIZE
+    add.d XX, XX, INCX
+    fld.d b3, Y, 0 * SIZE
+    fst.d a3, Y, 0 * SIZE
+    fld.d a4, X, 0 * SIZE
+    add.d X, X, INCX
+    fst.d b4, XX, 0 * SIZE
+    add.d XX, XX, INCX
+    fld.d b4, Y, 0 * SIZE
+    fst.d a4, Y, 0 * SIZE
+    add.d Y, Y, INCY
+    fst.d b1, XX, 0 * SIZE
+    add.d XX, XX, INCX
+    fst.d b2, XX, 0 * SIZE
+    add.d XX, XX, INCX
+    fst.d b3, XX, 0 * SIZE
+    add.d XX, XX, INCX
+    fst.d b4, XX, 0 * SIZE
+    add.d XX, XX, INCX
+    addi.d  I, I, -1
+    blt $r0, I, .L222
+    .align 3
+
+.L223:
+    andi I, N, 7
+    bge $r0, I, .L999
+    .align 3
+
+.L224:
+    fld.d $f12, X, 0 * SIZE
+    fld.d $f14, Y, 0 * SIZE
+    addi.d I, I, -1
+    fst.d $f12, Y, 0 * SIZE
+    fst.d $f14, X, 0 * SIZE
+    add.d  X, X, INCX
+    add.d  Y, Y, INCY
+    blt $r0, I, .L224
+    .align 3
+
+.L999:
+    move $r4, $r12
+    jirl $r0, $r1, 0x0
+    .align 3
+
+    EPILOGUE
diff --git a/kernel/loongarch64/sswap_lasx.S b/kernel/loongarch64/sswap_lasx.S
new file mode 100644
index 000000000..7184eff45
--- /dev/null
+++ b/kernel/loongarch64/sswap_lasx.S
@@ -0,0 +1,286 @@
+#define ASSEMBLER
+
+#include "common.h"
+#define N      $r4
+#define X      $r7
+#define INCX   $r8
+#define Y      $r9
+#define INCY   $r10
+
+#define I      $r17
+#define TEMP   $r18
+#define XX     $r5
+#define YY     $r6
+#define t1     $r14
+#define t2     $r15
+#define t3     $r16
+#define t4     $r19
+#define a1     $f12
+#define a2     $f13
+#define a3     $f14
+#define a4     $f15
+#define b1     $f16
+#define b2     $f17
+#define b3     $f18
+#define b4     $f19
+#define VX0    $xr12
+#define VX1    $xr13
+#define VX2    $xr14
+#define VX3    $xr15
+
+
+    PROLOGUE
+    bge $r0, N, .L999
+    li.d TEMP, 1
+    slli.d  TEMP, TEMP, BASE_SHIFT
+    slli.d  INCX, INCX, BASE_SHIFT
+    slli.d  INCY, INCY, BASE_SHIFT
+    srai.d I, N, 3
+    bne INCX, TEMP, .L20
+    bne INCY, TEMP, .L12 // INCX==1 and INCY!=1
+    b .L11  // INCX==1 and INCY==1
+.L20:
+    bne INCY, TEMP, .L22 // INCX!=1 and INCY!=1
+    b .L21 // INCX!=1 and INCY==1
+
+.L11:
+    bge $r0, I, .L112
+    .align 3
+
+.L111:
+    xvld VX0, X, 0 * SIZE
+    xvld VX2, Y, 0 * SIZE
+    addi.d  I, I, -1
+    xvst VX2, X, 0 * SIZE
+    xvst VX0, Y, 0 * SIZE
+    addi.d X, X, 8 * SIZE
+    addi.d Y, Y, 8 * SIZE
+    blt $r0, I, .L111
+    .align 3
+
+.L112:
+    andi I, N, 7
+    bge $r0, I, .L999
+    .align 3
+
+.L113:
+    fld.s $f12, X, 0 * SIZE
+    fld.s $f14, Y, 0 * SIZE
+    addi.d I, I, -1
+    fst.s $f12, Y, 0 * SIZE
+    fst.s $f14, X, 0 * SIZE
+    addi.d  X, X, SIZE
+    addi.d  Y, Y, SIZE
+    blt $r0, I, .L113
+    b .L999
+    .align 3
+
+.L12: // INCX==1 and INCY!=1
+    bge $r0, I, .L122
+    .align 3
+
+.L121:
+    xvld VX0, X, 0 * SIZE
+    ld.w t1, Y, 0 * SIZE
+    xvstelm.w VX0, Y, 0, 0
+    add.d Y, Y, INCY
+    ld.w t2, Y, 0 * SIZE
+    xvstelm.w VX0, Y, 0, 1
+    add.d Y, Y, INCY
+    ld.w t3, Y, 0 * SIZE
+    xvstelm.w VX0, Y, 0, 2
+    add.d Y, Y, INCY
+    ld.w t4, Y, 0 * SIZE
+    xvstelm.w VX0, Y, 0, 3
+    xvinsgr2vr.w VX2, t1, 0
+    xvinsgr2vr.w VX2, t2, 1
+    xvinsgr2vr.w VX2, t3, 2
+    xvinsgr2vr.w VX2, t4, 3
+    add.d Y, Y, INCY
+    ld.w t1, Y, 0 * SIZE
+    xvstelm.w VX0, Y, 0, 4
+    add.d Y, Y, INCY
+    ld.w t2, Y, 0 * SIZE
+    xvstelm.w VX0, Y, 0, 5
+    add.d Y, Y, INCY
+    ld.w t3, Y, 0 * SIZE
+    xvstelm.w VX0, Y, 0, 6
+    add.d Y, Y, INCY
+    ld.w t4, Y, 0 * SIZE
+    xvstelm.w VX0, Y, 0, 7
+    xvinsgr2vr.w VX2, t1, 4
+    xvinsgr2vr.w VX2, t2, 5
+    xvinsgr2vr.w VX2, t3, 6
+    xvinsgr2vr.w VX2, t4, 7
+    add.d Y, Y, INCY
+    xvst VX2, X, 0 * SIZE
+    addi.d X, X, 8 * SIZE
+    addi.d  I, I, -1
+    blt $r0, I, .L121
+    .align 3
+
+.L122:
+    andi I, N, 7
+    bge $r0, I, .L999
+    .align 3
+
+.L123:
+    fld.s $f12, X, 0 * SIZE
+    fld.s $f14, Y, 0 * SIZE
+    addi.d I, I, -1
+    fst.s $f12, Y, 0 * SIZE
+    fst.s $f14, X, 0 * SIZE
+    addi.d  X, X, SIZE
+    add.d  Y, Y, INCY
+    blt $r0, I, .L123
+    b .L999
+    .align 3
+
+.L21:
+    bge $r0, I, .L212
+    .align 3
+
+.L211:
+    xvld VX2, Y, 0 * SIZE
+    ld.w t1, X, 0 * SIZE
+    xvstelm.w VX2, X, 0, 0
+    add.d X, X, INCX
+    ld.w t2, X, 0 * SIZE
+    xvstelm.w VX2, X, 0, 1
+    add.d X, X, INCY
+    ld.w t3, X, 0 * SIZE
+    xvstelm.w VX2, X, 0, 2
+    add.d X, X, INCX
+    ld.w t4, X, 0 * SIZE
+    xvstelm.w VX2, X, 0, 3
+    xvinsgr2vr.w VX0, t1, 0
+    xvinsgr2vr.w VX0, t2, 1
+    xvinsgr2vr.w VX0, t3, 2
+    xvinsgr2vr.w VX0, t4, 3
+    add.d X, X, INCX
+    ld.w t1, X, 0 * SIZE
+    xvstelm.w VX2, X, 0, 4
+    add.d X, X, INCY
+    ld.w t2, X, 0 * SIZE
+    xvstelm.w VX2, X, 0, 5
+    add.d X, X, INCX
+    ld.w t3, X, 0 * SIZE
+    xvstelm.w VX2, X, 0, 6
+    add.d X, X, INCX
+    ld.w t4, X, 0 * SIZE
+    xvstelm.w VX2, X, 0, 7
+    xvinsgr2vr.w VX0, t1, 4
+    xvinsgr2vr.w VX0, t2, 5
+    xvinsgr2vr.w VX0, t3, 6
+    xvinsgr2vr.w VX0, t4, 7
+    add.d X, X, INCX
+    xvst VX1, Y, 0 * SIZE
+    addi.d Y, Y, 8 * SIZE
+    addi.d  I, I, -1
+    blt $r0, I, .L211
+    .align 3
+
+.L212:
+    andi I, N, 7
+    bge $r0, I, .L999
+    .align 3
+
+.L213:
+    fld.s $f12, X, 0 * SIZE
+    fld.s $f14, Y, 0 * SIZE
+    addi.d I, I, -1
+    fst.s $f12, Y, 0 * SIZE
+    fst.s $f14, X, 0 * SIZE
+    add.d  X, X, INCX
+    addi.d  Y, Y, SIZE
+    blt $r0, I, .L213
+    b .L999
+    .align 3
+
+.L22:
+    bge $r0, I, .L223
+    .align 3
+    move XX, X
+
+.L222:
+    fld.s a1, X, 0 * SIZE
+    add.d X, X, INCX
+    fld.s a2, X, 0 * SIZE
+    add.d X, X, INCX
+    fld.s a3, X, 0 * SIZE
+    add.d X, X, INCX
+    fld.s a4, X, 0 * SIZE
+    add.d X, X, INCX
+    fld.s b1, Y, 0 * SIZE
+    fst.s a1, Y, 0 * SIZE
+    add.d Y, Y, INCY
+    fld.s b2, Y, 0 * SIZE
+    fst.s a2, Y, 0 * SIZE
+    add.d Y, Y, INCY
+    fld.s b3, Y, 0 * SIZE
+    fst.s a3, Y, 0 * SIZE
+    add.d Y, Y, INCY
+    fld.s b4, Y, 0 * SIZE
+    fst.s a4, Y, 0 * SIZE
+    add.d Y, Y, INCY
+    fld.s a1, X, 0 * SIZE
+    add.d X, X, INCX
+    fst.s b1, XX, 0 * SIZE
+    add.d XX, XX, INCX
+    fld.s b1, Y, 0 * SIZE
+    fst.s a1, Y, 0 * SIZE
+    add.d Y, Y, INCY
+    fld.s a2, X, 0 * SIZE
+    add.d X, X, INCX
+    fst.s b2, XX, 0 * SIZE
+    add.d XX, XX, INCX
+    fld.s b2, Y, 0 * SIZE
+    fst.s a2, Y, 0 * SIZE
+    add.d Y, Y, INCY
+    fld.s a3, X, 0 * SIZE
+    add.d X, X, INCX
+    fst.s b3, XX, 0 * SIZE
+    add.d XX, XX, INCX
+    fld.s b3, Y, 0 * SIZE
+    fst.s a3, Y, 0 * SIZE
+    fld.s a4, X, 0 * SIZE
+    add.d X, X, INCX
+    fst.s b4, XX, 0 * SIZE
+    add.d XX, XX, INCX
+    fld.s b4, Y, 0 * SIZE
+    fst.s a4, Y, 0 * SIZE
+    add.d Y, Y, INCY
+    fst.s b1, XX, 0 * SIZE
+    add.d XX, XX, INCX
+    fst.s b2, XX, 0 * SIZE
+    add.d XX, XX, INCX
+    fst.s b3, XX, 0 * SIZE
+    add.d XX, XX, INCX
+    fst.s b4, XX, 0 * SIZE
+    add.d XX, XX, INCX
+    addi.d  I, I, -1
+    blt $r0, I, .L222
+    .align 3
+
+.L223:
+    andi I, N, 7
+    bge $r0, I, .L999
+    .align 3
+
+.L224:
+    fld.s $f12, X, 0 * SIZE
+    fld.s $f14, Y, 0 * SIZE
+    addi.d I, I, -1
+    fst.s $f12, Y, 0 * SIZE
+    fst.s $f14, X, 0 * SIZE
+    add.d  X, X, INCX
+    add.d  Y, Y, INCY
+    blt $r0, I, .L224
+    .align 3
+
+.L999:
+    move $r4, $r12
+    jirl $r0, $r1, 0x0
+    .align 3
+
+    EPILOGUE
diff --git a/kernel/loongarch64/sswap_lsx.S b/kernel/loongarch64/sswap_lsx.S
new file mode 100644
index 000000000..4f19a8024
--- /dev/null
+++ b/kernel/loongarch64/sswap_lsx.S
@@ -0,0 +1,294 @@
+#define ASSEMBLER
+
+#include "common.h"
+#define N      $r4
+#define X      $r7
+#define INCX   $r8
+#define Y      $r9
+#define INCY   $r10
+
+#define I      $r17
+#define TEMP   $r18
+#define XX     $r5
+#define YY     $r6
+#define t1     $r14
+#define t2     $r15
+#define t3     $r16
+#define t4     $r19
+#define a1     $f12
+#define a2     $f13
+#define a3     $f14
+#define a4     $f15
+#define b1     $f16
+#define b2     $f17
+#define b3     $f18
+#define b4     $f19
+#define VX0    $vr12
+#define VX1    $vr13
+#define VX2    $vr14
+#define VX3    $vr15
+
+
+    PROLOGUE
+    bge $r0, N, .L999
+    li.d TEMP, 1
+    slli.d  TEMP, TEMP, BASE_SHIFT
+    slli.d  INCX, INCX, BASE_SHIFT
+    slli.d  INCY, INCY, BASE_SHIFT
+    srai.d I, N, 3
+    bne INCX, TEMP, .L20
+    bne INCY, TEMP, .L12 // INCX==1 and INCY!=1
+    b .L11  // INCX==1 and INCY==1
+.L20:
+    bne INCY, TEMP, .L22 // INCX!=1 and INCY!=1
+    b .L21 // INCX!=1 and INCY==1
+
+.L11:
+    bge $r0, I, .L112
+    .align 3
+
+.L111:
+    vld VX0, X, 0 * SIZE
+    vld VX1, X, 4 * SIZE
+    vld VX2, Y, 0 * SIZE
+    vld VX3, Y, 4 * SIZE
+    addi.d  I, I, -1
+    vst VX2, X, 0 * SIZE
+    vst VX3, X, 4 * SIZE
+    vst VX0, Y, 0 * SIZE
+    vst VX1, Y, 4 * SIZE
+    addi.d X, X, 8 * SIZE
+    addi.d Y, Y, 8 * SIZE
+    blt $r0, I, .L111
+    .align 3
+
+.L112:
+    andi I, N, 7
+    bge $r0, I, .L999
+    .align 3
+
+.L113:
+    fld.s $f12, X, 0 * SIZE
+    fld.s $f14, Y, 0 * SIZE
+    addi.d I, I, -1
+    fst.s $f12, Y, 0 * SIZE
+    fst.s $f14, X, 0 * SIZE
+    addi.d  X, X, SIZE
+    addi.d  Y, Y, SIZE
+    blt $r0, I, .L113
+    b .L999
+    .align 3
+
+.L12: // INCX==1 and INCY!=1
+    bge $r0, I, .L122
+    .align 3
+
+.L121:
+    vld VX0, X, 0 * SIZE
+    ld.w t1, Y, 0 * SIZE
+    vstelm.w VX0, Y, 0, 0
+    add.d Y, Y, INCY
+    ld.w t2, Y, 0 * SIZE
+    vstelm.w VX0, Y, 0, 1
+    add.d Y, Y, INCY
+    ld.w t3, Y, 0 * SIZE
+    vstelm.w VX0, Y, 0, 2
+    add.d Y, Y, INCY
+    ld.w t4, Y, 0 * SIZE
+    vstelm.w VX0, Y, 0, 3
+    vinsgr2vr.w VX2, t1, 0
+    vinsgr2vr.w VX2, t2, 1
+    vinsgr2vr.w VX2, t3, 2
+    vinsgr2vr.w VX2, t4, 3
+    add.d Y, Y, INCY
+    vst VX2, X, 0 * SIZE
+    vld VX1, X, 4 * SIZE
+    ld.w t1, Y, 0 * SIZE
+    vstelm.w VX1, Y, 0, 0
+    add.d Y, Y, INCY
+    ld.w t2, Y, 0 * SIZE
+    vstelm.w VX1, Y, 0, 1
+    add.d Y, Y, INCY
+    ld.w t3, Y, 0 * SIZE
+    vstelm.w VX1, Y, 0, 2
+    add.d Y, Y, INCY
+    ld.w t4, Y, 0 * SIZE
+    vstelm.w VX1, Y, 0, 3
+    vinsgr2vr.w VX3, t1, 0
+    vinsgr2vr.w VX3, t2, 1
+    vinsgr2vr.w VX3, t3, 2
+    vinsgr2vr.w VX3, t4, 3
+    add.d Y, Y, INCY
+    vst VX3, X, 4 * SIZE
+    addi.d X, X, 8 * SIZE
+    addi.d  I, I, -1
+    blt $r0, I, .L121
+    .align 3
+
+.L122:
+    andi I, N, 7
+    bge $r0, I, .L999
+    .align 3
+
+.L123:
+    fld.s $f12, X, 0 * SIZE
+    fld.s $f14, Y, 0 * SIZE
+    addi.d I, I, -1
+    fst.s $f12, Y, 0 * SIZE
+    fst.s $f14, X, 0 * SIZE
+    addi.d  X, X, SIZE
+    add.d  Y, Y, INCY
+    blt $r0, I, .L123
+    b .L999
+    .align 3
+
+.L21:// INCX!=1 and INCY==1
+    bge $r0, I, .L212
+    .align 3
+
+.L211:
+    vld VX2, Y, 0 * SIZE
+    ld.w t1, X, 0 * SIZE
+    vstelm.w VX2, X, 0, 0
+    add.d X, X, INCX
+    ld.w t2, X, 0 * SIZE
+    vstelm.w VX2, X, 0, 1
+    add.d X, X, INCY
+    ld.w t3, X, 0 * SIZE
+    vstelm.w VX2, X, 0, 2
+    add.d X, X, INCX
+    ld.w t4, X, 0 * SIZE
+    vstelm.w VX2, X, 0, 3
+    vinsgr2vr.w VX0, t1, 0
+    vinsgr2vr.w VX0, t2, 1
+    vinsgr2vr.w VX0, t3, 2
+    vinsgr2vr.w VX0, t4, 3
+    add.d X, X, INCX
+    vst VX0, Y, 0 * SIZE
+    vld VX3, Y, 4 * SIZE
+    ld.w t1, X, 0 * SIZE
+    vstelm.w VX3, X, 0, 0
+    add.d X, X, INCY
+    ld.w t2, X, 0 * SIZE
+    vstelm.w VX3, X, 0, 1
+    add.d X, X, INCX
+    ld.w t3, X, 0 * SIZE
+    vstelm.w VX3, X, 0, 2
+    add.d X, X, INCX
+    ld.w t4, X, 0 * SIZE
+    vstelm.w VX3, X, 0, 3
+    vinsgr2vr.w VX1, t1, 0
+    vinsgr2vr.w VX1, t2, 1
+    vinsgr2vr.w VX1, t3, 2
+    vinsgr2vr.w VX1, t4, 3
+    add.d X, X, INCX
+    vst VX1, Y, 0 * SIZE
+    addi.d Y, Y, 8 * SIZE
+    addi.d  I, I, -1
+    blt $r0, I, .L211
+    .align 3
+
+.L212:
+    andi I, N, 7
+    bge $r0, I, .L999
+    .align 3
+
+.L213:
+    fld.s $f12, X, 0 * SIZE
+    fld.s $f14, Y, 0 * SIZE
+    addi.d I, I, -1
+    fst.s $f12, Y, 0 * SIZE
+    fst.s $f14, X, 0 * SIZE
+    add.d  X, X, INCX
+    addi.d  Y, Y, SIZE
+    blt $r0, I, .L213
+    b .L999
+    .align 3
+
+.L22:
+    bge $r0, I, .L223
+    .align 3
+    move XX, X
+
+.L222:
+    fld.s a1, X, 0 * SIZE
+    add.d X, X, INCX
+    fld.s a2, X, 0 * SIZE
+    add.d X, X, INCX
+    fld.s a3, X, 0 * SIZE
+    add.d X, X, INCX
+    fld.s a4, X, 0 * SIZE
+    add.d X, X, INCX
+    fld.s b1, Y, 0 * SIZE
+    fst.s a1, Y, 0 * SIZE
+    add.d Y, Y, INCY
+    fld.s b2, Y, 0 * SIZE
+    fst.s a2, Y, 0 * SIZE
+    add.d Y, Y, INCY
+    fld.s b3, Y, 0 * SIZE
+    fst.s a3, Y, 0 * SIZE
+    add.d Y, Y, INCY
+    fld.s b4, Y, 0 * SIZE
+    fst.s a4, Y, 0 * SIZE
+    add.d Y, Y, INCY
+    fld.s a1, X, 0 * SIZE
+    add.d X, X, INCX
+    fst.s b1, XX, 0 * SIZE
+    add.d XX, XX, INCX
+    fld.s b1, Y, 0 * SIZE
+    fst.s a1, Y, 0 * SIZE
+    add.d Y, Y, INCY
+    fld.s a2, X, 0 * SIZE
+    add.d X, X, INCX
+    fst.s b2, XX, 0 * SIZE
+    add.d XX, XX, INCX
+    fld.s b2, Y, 0 * SIZE
+    fst.s a2, Y, 0 * SIZE
+    add.d Y, Y, INCY
+    fld.s a3, X, 0 * SIZE
+    add.d X, X, INCX
+    fst.s b3, XX, 0 * SIZE
+    add.d XX, XX, INCX
+    fld.s b3, Y, 0 * SIZE
+    fst.s a3, Y, 0 * SIZE
+    fld.s a4, X, 0 * SIZE
+    add.d X, X, INCX
+    fst.s b4, XX, 0 * SIZE
+    add.d XX, XX, INCX
+    fld.s b4, Y, 0 * SIZE
+    fst.s a4, Y, 0 * SIZE
+    add.d Y, Y, INCY
+    fst.s b1, XX, 0 * SIZE
+    add.d XX, XX, INCX
+    fst.s b2, XX, 0 * SIZE
+    add.d XX, XX, INCX
+    fst.s b3, XX, 0 * SIZE
+    add.d XX, XX, INCX
+    fst.s b4, XX, 0 * SIZE
+    add.d XX, XX, INCX
+    addi.d  I, I, -1
+    blt $r0, I, .L222
+    .align 3
+
+.L223:
+    andi I, N, 7
+    bge $r0, I, .L999
+    .align 3
+
+.L224:
+    fld.s $f12, X, 0 * SIZE
+    fld.s $f14, Y, 0 * SIZE
+    addi.d I, I, -1
+    fst.s $f12, Y, 0 * SIZE
+    fst.s $f14, X, 0 * SIZE
+    add.d  X, X, INCX
+    add.d  Y, Y, INCY
+    blt $r0, I, .L224
+    .align 3
+
+.L999:
+    move $r4, $r12
+    jirl $r0, $r1, 0x0
+    .align 3
+
+    EPILOGUE