revert ggml-org#15953 for now as it breaks kokoro

Author: LostRuins

ggml/src/ggml-cpu/ops.cpp

@@ -3467,27 +3467,31 @@ static void ggml_compute_forward_norm_f32(
 
     GGML_ASSERT(eps >= 0.0f);
 
+    // kcpp: reverted back to non accelerated, because it was breaking kokoro
     for (int64_t i03 = 0; i03 < ne03; i03++) {
         for (int64_t i02 = 0; i02 < ne02; i02++) {
             for (int64_t i01 = ith; i01 < ne01; i01 += nth) {
                 const float * x = (float *) ((char *) src0->data + i01*nb01 + i02*nb02 + i03*nb03);
 
-                float sum = 0.0;
-                ggml_vec_sum_f32(ne00, &sum, x);
+                ggml_float sum = 0.0;
+                for (int64_t i00 = 0; i00 < ne00; i00++) {
+                    sum += (ggml_float)x[i00];
+                }
+
                 float mean = sum/ne00;
 
                 float * y = (float *) ((char *) dst->data + i01*nb1 + i02*nb2 + i03*nb3);
-                float variance = 0;
 
-#ifdef GGML_USE_ACCELERATE
-                mean = -mean;
-                vDSP_vsadd(x, 1, &mean, y, 1, ne00);
-                vDSP_measqv(y, 1, &variance, ne00);
-#else
-                variance = ggml_vec_cvar_f32(ne00, y, x, mean);
-#endif //GGML_USE_ACCELERATE
+                ggml_float sum2 = 0.0;
+                for (int64_t i00 = 0; i00 < ne00; i00++) {
+                    float v = x[i00] - mean;
+                    y[i00] = v;
+                    sum2 += (ggml_float)(v*v);
+                }
 
+                float variance = sum2/ne00;
                 const float scale = 1.0f/sqrtf(variance + eps);
+
                 ggml_vec_scale_f32(ne00, y, scale);
             }
         }