commit 74921eee924625426429044decefe3673561b174
author Michael Tyler <michael.tyler@arm.com> Wed Apr 12 17:43:17 2023 +0100
committer michael.tyler <michael.tyler@arm.com> Mon Jun 05 15:57:58 2023 +0000
tree 654da1a95e3d42d6af8ad1ff27bb40d77b1fd8c5
parent df5d9878008be9b60586df97ebfff197abb5195e

Update CPU kernel implementations and guard directives

Resolves COMPMID-6023

Change-Id: I868975d14c4f98af6716726feda22405a6a4c891
Signed-off-by: Michael Tyler <michael.tyler@arm.com>
Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/9686
Tested-by: Arm Jenkins <bsgcomp@arm.com>
Reviewed-by: Viet-Hoa Do <viet-hoa.do@arm.com>
Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
Benchmark: Arm Jenkins <bsgcomp@arm.com>

src/core/NEON/kernels/arm_gemm/indirect-interleaves/sme_interleave1VL_block4_u8_u8_summing.hpp

diff --git a/src/core/NEON/kernels/arm_gemm/indirect-interleaves/sme_interleave1VL_block4_u8_u8_summing.hpp b/src/core/NEON/kernels/arm_gemm/indirect-interleaves/sme_interleave1VL_block4_u8_u8_summing.hpp
index 646db0c..daf2d3a 100644
--- a/src/core/NEON/kernels/arm_gemm/indirect-interleaves/sme_interleave1VL_block4_u8_u8_summing.hpp
+++ b/src/core/NEON/kernels/arm_gemm/indirect-interleaves/sme_interleave1VL_block4_u8_u8_summing.hpp
@@ -22,7 +22,7 @@
  * SOFTWARE.
  */
 
-#if defined(__ARM_FEATURE_SVE)
+#if defined(ARM_COMPUTE_ENABLE_SME)
 
 template <>
 void interleave_block<1, 4, VLType::SME, true>(
@@ -200,12 +200,12 @@
       "10:"  // K loop: Tails: Even: First
       ".inst 0x25306d20  // psel p0.s, p11.s/Z, p9.s[w12]\n"
       ".inst 0xe0bf8300  // st1w { za0v.s[x12] }, p0/Z, [x24, XZR, LSL #2]\n"
-      "ldr x22, [x23, #0x0]\n"
+      "ldr x20, [x23, #0x0]\n"
       ".inst 0xc0828810  // mova z16.s, p2/M, za0v.s[x12]\n"
       "add x12, x12, #0x1\n"
       ".inst 0x25356140  // psel p0.b, p8.b/Z, p10.b[w13, #2]\n"
       "udot z17.s, z16.b, z18.b\n"
-      ".inst 0xe01922c2  // ld1b { za0h.b[x13, #2] }, p0/Z, [x22, x25]\n"
+      ".inst 0xe0192282  // ld1b { za0h.b[x13, #2] }, p0/Z, [x20, x25]\n"
       "cmp x12, x9\n"
       "add x23, x23, #0x8\n"
       "addvl x24, x24, #1\n"
@@ -225,7 +225,7 @@
       "addvl x24, x24, #1\n"
       "add x20, x20, #0x4\n"
       "blt 11b\n"
-      "whilelt p9.b, x28, %x[width]\n"
+      "whilelt p8.b, x28, %x[width]\n"
       "b 14f\n"
       "12:"  // K loop: Tails: Odd
       "mov x12, #0x0\n"
@@ -249,4 +249,4 @@
     );
 }
 
-#endif  // defined(__ARM_FEATURE_SVE)
+#endif  // defined(ARM_COMPUTE_ENABLE_SME)