diff --git a/src/core/NEON/kernels/arm_gemm/transforms/sme_transpose_interleave_4VL.hpp b/src/core/NEON/kernels/arm_gemm/transforms/sme_transpose_interleave_4VL.hpp
index 7b783c3..362bebb 100644
--- a/src/core/NEON/kernels/arm_gemm/transforms/sme_transpose_interleave_4VL.hpp
+++ b/src/core/NEON/kernels/arm_gemm/transforms/sme_transpose_interleave_4VL.hpp
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2022 Arm Limited.
+ * Copyright (c) 2022-2023 Arm Limited.
  *
  * SPDX-License-Identifier: MIT
  *
@@ -10,16 +10,16 @@
  * sell copies of the Software, and to permit persons to whom the Software is
  * furnished to do so, subject to the following conditions:
  *
- * The above copyright notice and this permission notice shall be included in
- * all copies or substantial portions of the Software.
+ * The above copyright notice and this permission notice shall be included in all
+ * copies or substantial portions of the Software.
  *
  * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
  * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
  * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
  * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
- * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
- * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
- * IN THE SOFTWARE.
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+ * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+ * SOFTWARE.
  */
 
 #pragma once
@@ -39,64 +39,64 @@
       "ptrue p4.b\n"
       "blt 4f\n"
       "1:"  // Main row loop: Head
-      "mov x25, %x[in]\n"
+      "mov x26, %x[in]\n"
+      "add x25, x26, %x[in_stride]\n"
       "add x24, x25, %x[in_stride]\n"
       "add x23, x24, %x[in_stride]\n"
-      "add x22, x23, %x[in_stride]\n"
-      "add %x[in], x22, %x[in_stride]\n"
-      "mov x21, %x[out]\n"
+      "add %x[in], x23, %x[in_stride]\n"
+      "mov x22, %x[out]\n"
       "sub %x[height], %x[height], #0x4\n"
-      "mov x20, %x[width]\n"
+      "mov x21, %x[width]\n"
       "2:"  // Main row loop: Column loop
-      "mov x19, x20\n"
-      "whilelt p3.h, XZR, x19\n"
-      "ld1h { z31.h }, p3/Z, [x25]\n"
-      "dech x19\n"
-      "whilelt p2.h, XZR, x19\n"
-      "ld1h { z30.h }, p2/Z, [x25, #1, MUL VL]\n"
-      "dech x19\n"
-      "whilelt p1.h, XZR, x19\n"
-      "ld1h { z29.h }, p1/Z, [x25, #2, MUL VL]\n"
-      "dech x19\n"
-      "whilelt p0.h, XZR, x19\n"
-      "ld1h { z28.h }, p0/Z, [x25, #3, MUL VL]\n"
-      "mov x19, x21\n"
-      "dech x20, ALL, MUL #4\n"
-      "ld1h { z27.h }, p3/Z, [x24]\n"
-      "ld1h { z26.h }, p2/Z, [x24, #1, MUL VL]\n"
-      "cmp x20, #0x0\n"
+      "mov x20, x21\n"
+      "whilelt p3.h, XZR, x20\n"
+      "ld1h { z31.h }, p3/Z, [x26]\n"
+      "dech x20\n"
+      "whilelt p2.h, XZR, x20\n"
+      "ld1h { z30.h }, p2/Z, [x26, #1, MUL VL]\n"
+      "dech x20\n"
+      "whilelt p1.h, XZR, x20\n"
+      "ld1h { z29.h }, p1/Z, [x26, #2, MUL VL]\n"
+      "dech x20\n"
+      "whilelt p0.h, XZR, x20\n"
+      "ld1h { z28.h }, p0/Z, [x26, #3, MUL VL]\n"
+      "mov x20, x22\n"
+      "dech x21, ALL, MUL #4\n"
+      "ld1h { z27.h }, p3/Z, [x25]\n"
+      "ld1h { z26.h }, p2/Z, [x25, #1, MUL VL]\n"
+      "cmp x21, #0x0\n"
+      "addvl x26, x26, #4\n"
+      "ld1h { z25.h }, p1/Z, [x25, #2, MUL VL]\n"
+      "add x22, x22, %x[out_stride]\n"
+      "ld1h { z24.h }, p0/Z, [x25, #3, MUL VL]\n"
       "addvl x25, x25, #4\n"
-      "ld1h { z25.h }, p1/Z, [x24, #2, MUL VL]\n"
-      "add x21, x21, %x[out_stride]\n"
-      "ld1h { z24.h }, p0/Z, [x24, #3, MUL VL]\n"
+      "ld1h { z23.h }, p3/Z, [x24]\n"
+      "ld1h { z22.h }, p2/Z, [x24, #1, MUL VL]\n"
+      "ld1h { z21.h }, p1/Z, [x24, #2, MUL VL]\n"
+      "ld1h { z20.h }, p0/Z, [x24, #3, MUL VL]\n"
       "addvl x24, x24, #4\n"
-      "ld1h { z23.h }, p3/Z, [x23]\n"
-      "ld1h { z22.h }, p2/Z, [x23, #1, MUL VL]\n"
-      "ld1h { z21.h }, p1/Z, [x23, #2, MUL VL]\n"
-      "ld1h { z20.h }, p0/Z, [x23, #3, MUL VL]\n"
+      "ld1h { z19.h }, p3/Z, [x23]\n"
+      "ld1h { z18.h }, p2/Z, [x23, #1, MUL VL]\n"
+      "ld1h { z17.h }, p1/Z, [x23, #2, MUL VL]\n"
+      "ld1h { z16.h }, p0/Z, [x23, #3, MUL VL]\n"
+      "st1h { z31.h }, p4, [x20]\n"
       "addvl x23, x23, #4\n"
-      "ld1h { z19.h }, p3/Z, [x22]\n"
-      "ld1h { z18.h }, p2/Z, [x22, #1, MUL VL]\n"
-      "ld1h { z17.h }, p1/Z, [x22, #2, MUL VL]\n"
-      "ld1h { z16.h }, p0/Z, [x22, #3, MUL VL]\n"
-      "st1h { z31.h }, p4, [x19]\n"
-      "addvl x22, x22, #4\n"
-      "st1h { z30.h }, p4, [x19, #1, MUL VL]\n"
-      "st1h { z29.h }, p4, [x19, #2, MUL VL]\n"
-      "st1h { z28.h }, p4, [x19, #3, MUL VL]\n"
-      "st1h { z27.h }, p4, [x19, #4, MUL VL]\n"
-      "st1h { z26.h }, p4, [x19, #5, MUL VL]\n"
-      "st1h { z25.h }, p4, [x19, #6, MUL VL]\n"
-      "st1h { z24.h }, p4, [x19, #7, MUL VL]\n"
-      "addvl x19, x19, #16\n"
-      "st1h { z23.h }, p4, [x19, #-8, MUL VL]\n"
-      "st1h { z22.h }, p4, [x19, #-7, MUL VL]\n"
-      "st1h { z21.h }, p4, [x19, #-6, MUL VL]\n"
-      "st1h { z20.h }, p4, [x19, #-5, MUL VL]\n"
-      "st1h { z19.h }, p4, [x19, #-4, MUL VL]\n"
-      "st1h { z18.h }, p4, [x19, #-3, MUL VL]\n"
-      "st1h { z17.h }, p4, [x19, #-2, MUL VL]\n"
-      "st1h { z16.h }, p4, [x19, #-1, MUL VL]\n"
+      "st1h { z30.h }, p4, [x20, #1, MUL VL]\n"
+      "st1h { z29.h }, p4, [x20, #2, MUL VL]\n"
+      "st1h { z28.h }, p4, [x20, #3, MUL VL]\n"
+      "st1h { z27.h }, p4, [x20, #4, MUL VL]\n"
+      "st1h { z26.h }, p4, [x20, #5, MUL VL]\n"
+      "st1h { z25.h }, p4, [x20, #6, MUL VL]\n"
+      "st1h { z24.h }, p4, [x20, #7, MUL VL]\n"
+      "addvl x20, x20, #16\n"
+      "st1h { z23.h }, p4, [x20, #-8, MUL VL]\n"
+      "st1h { z22.h }, p4, [x20, #-7, MUL VL]\n"
+      "st1h { z21.h }, p4, [x20, #-6, MUL VL]\n"
+      "st1h { z20.h }, p4, [x20, #-5, MUL VL]\n"
+      "st1h { z19.h }, p4, [x20, #-4, MUL VL]\n"
+      "st1h { z18.h }, p4, [x20, #-3, MUL VL]\n"
+      "st1h { z17.h }, p4, [x20, #-2, MUL VL]\n"
+      "st1h { z16.h }, p4, [x20, #-1, MUL VL]\n"
       "bgt 2b\n"
       "3:"  // Main row loop: Column loop skip
       "cmp %x[height], #0x4\n"
@@ -105,32 +105,32 @@
       "cbz %x[height], 8f\n"
       "4:"  // Main loop skip
       "5:"  // Tail row loop: Head
-      "mov x25, %x[in]\n"
-      "add %x[in], x25, %x[in_stride]\n"
-      "mov x21, %x[out]\n"
+      "mov x26, %x[in]\n"
+      "add %x[in], x26, %x[in_stride]\n"
+      "mov x22, %x[out]\n"
       "sub %x[height], %x[height], #0x1\n"
-      "mov x20, %x[width]\n"
+      "mov x21, %x[width]\n"
       "6:"  // Tail row loop: Column loop
-      "mov x19, x20\n"
-      "whilelt p0.h, XZR, x19\n"
-      "ld1h { z19.h }, p0/Z, [x25]\n"
-      "dech x19\n"
-      "whilelt p0.h, XZR, x19\n"
-      "ld1h { z18.h }, p0/Z, [x25, #1, MUL VL]\n"
-      "dech x19\n"
-      "whilelt p0.h, XZR, x19\n"
-      "ld1h { z17.h }, p0/Z, [x25, #2, MUL VL]\n"
-      "dech x19\n"
-      "dech x20, ALL, MUL #4\n"
-      "whilelt p0.h, XZR, x19\n"
-      "cmp x20, #0x0\n"
-      "ld1h { z16.h }, p0/Z, [x25, #3, MUL VL]\n"
-      "st1h { z19.h }, p4, [x21]\n"
-      "addvl x25, x25, #4\n"
-      "st1h { z18.h }, p4, [x21, #1, MUL VL]\n"
-      "st1h { z17.h }, p4, [x21, #2, MUL VL]\n"
-      "st1h { z16.h }, p4, [x21, #3, MUL VL]\n"
-      "add x21, x21, %x[out_stride]\n"
+      "mov x20, x21\n"
+      "whilelt p0.h, XZR, x20\n"
+      "ld1h { z19.h }, p0/Z, [x26]\n"
+      "dech x20\n"
+      "whilelt p0.h, XZR, x20\n"
+      "ld1h { z18.h }, p0/Z, [x26, #1, MUL VL]\n"
+      "dech x20\n"
+      "whilelt p0.h, XZR, x20\n"
+      "ld1h { z17.h }, p0/Z, [x26, #2, MUL VL]\n"
+      "dech x20\n"
+      "dech x21, ALL, MUL #4\n"
+      "whilelt p0.h, XZR, x20\n"
+      "cmp x21, #0x0\n"
+      "ld1h { z16.h }, p0/Z, [x26, #3, MUL VL]\n"
+      "st1h { z19.h }, p4, [x22]\n"
+      "addvl x26, x26, #4\n"
+      "st1h { z18.h }, p4, [x22, #1, MUL VL]\n"
+      "st1h { z17.h }, p4, [x22, #2, MUL VL]\n"
+      "st1h { z16.h }, p4, [x22, #3, MUL VL]\n"
+      "add x22, x22, %x[out_stride]\n"
       "bgt 6b\n"
       "7:"  // Tail row loop: Column loop skip
       "cmp %x[height], #0x1\n"
@@ -140,7 +140,7 @@
       ".inst 0xd503467f  // SMSTOP\n"
       : [height] "+&r" (height), [in] "+&r" (in), [out] "+&r" (out)
       : [in_stride] "r" (in_stride), [out_stride] "r" (out_stride), [width] "r" (width)
-      : "cc", "memory", "p0", "p1", "p2", "p3", "p4", "p5", "p6", "p7", "p8", "p9", "p10", "p11", "p12", "p13", "p14", "p15", "x19", "x20", "x21", "x22", "x23", "x24", "x25", "z0", "z1", "z2", "z3", "z4", "z5", "z6", "z7", "z8", "z9", "z10", "z11", "z12", "z13", "z14", "z15", "z16", "z17", "z18", "z19", "z20", "z21", "z22", "z23", "z24", "z25", "z26", "z27", "z28", "z29", "z30", "z31"
+      : "cc", "memory", "p0", "p1", "p2", "p3", "p4", "p5", "p6", "p7", "p8", "p9", "p10", "p11", "p12", "p13", "p14", "p15", "x20", "x21", "x22", "x23", "x24", "x25", "x26", "z0", "z1", "z2", "z3", "z4", "z5", "z6", "z7", "z8", "z9", "z10", "z11", "z12", "z13", "z14", "z15", "z16", "z17", "z18", "z19", "z20", "z21", "z22", "z23", "z24", "z25", "z26", "z27", "z28", "z29", "z30", "z31"
     );
 }