cpu: aarch64: allow blocked weight format for primitive creation

with weights pre-packing enabled in torch.compile(),
the weights come already reorderd and in oneDNN format,
so, allowing format_kind::blocked as one of the supported
formats for acl inner product primitive.

src/cpu/aarch64/acl_inner_product.hpp

@@ -93,20 +93,25 @@ struct acl_inner_product_fwd_t : public primitive_t {
 
         status_t init(engine_t *engine) {
             using namespace data_type;
+            const format_kind_t weights_format_kind_received
+                    = weights_md_.format_kind;
             const bool is_fp16_ok = expect_data_types(f16, f16, f16, f16, undef)
                     && attr()->has_default_values(
                             primitive_attr_t::skip_mask_t::post_ops, f16);
             const bool is_fp32_ok = expect_data_types(f32, f32, f32, f32, undef)
                     && attr()->has_default_values(
                             primitive_attr_t::skip_mask_t::post_ops, f32);
+            const bool is_weights_md_format_ok
+                    = utils::one_of(weights_format_kind_received,
+                            format_kind::any, format_kind::blocked);
             const bool ok = is_fwd() && !has_zero_dim_memory()
                     && utils::one_of(true, is_fp16_ok, is_fp32_ok)
-                    && weights_md_.format_kind == format_kind::any
-                    && set_default_params() == status::success;
+                    && is_weights_md_format_ok
+                    && set_default_params(true) == status::success;
 
             if (!ok) return status::unimplemented;
 
-            CHECK(init_conf_ip(engine));
+            CHECK(init_conf_ip(engine, weights_format_kind_received));
 
             return status::success;
         }
@@ -115,7 +120,8 @@ struct acl_inner_product_fwd_t : public primitive_t {
 
         acl_post_ops_t post_ops;
 
-        status_t init_conf_ip(engine_t *engine) {
+        status_t init_conf_ip(
+                engine_t *engine, format_kind_t weights_format_kind_received) {
 
             ACL_CHECK_SUPPORT(src_md()->ndims != weights_md()->ndims,
                     "source and weights dimensions must match");
@@ -257,10 +263,19 @@ struct acl_inner_product_fwd_t : public primitive_t {
                     return status::unimplemented;
             }
 
+            const memory_desc_t weights_md_received = weights_md_;
             acl_utils::reorder_to_weight_format(aip.wei_tensor_info,
                     weights_md_, expected_weight_format, inner_dim, o_dim,
                     remaining_dims, {});
 
+            ACL_CHECK_SUPPORT(
+                    (weights_format_kind_received == format_kind::blocked)
+                            && !(dnnl_memory_desc_equal(
+                                    &weights_md_received, &weights_md_)),
+                    "specific blocked format not supported by ACL, use "
+                    "format_kind_t::any to find a supported blocked format for "
+                    "your platform");
+
             // clang-format off
 
             // Validate fully connected layer manually to check for return status