responseData.addMetaData(FILE_ATTRIBUTE_VIEW, aclFileAttributeView);
                responseData.addMetaData(FS_FILE_GROUPS,
                        aclFileAttributeView.getAcl().stream().map(acl -> acl.principal().getName()).toArray(n -> new String[n]));
                return aclFileAttributeView;
            }

        'details.')
  config_info_line('mkl', 'Build with MKL support.')
  config_info_line(
      'mkl_aarch64',
      'Build with oneDNN and Compute Library for the Arm Architecture (ACL).')
  config_info_line('monolithic', 'Config for mostly static monolithic build.')
  config_info_line('numa', 'Build with NUMA support.')
  config_info_line(
      'dynamic_kernels',

common:mkl_threadpool --define=tensorflow_mkldnn_contraction_kernel=0
common:mkl_threadpool --define=build_with_mkl_opensource=true
common:mkl_threadpool -c opt

# Config setting to build oneDNN with Compute Library for the Arm Architecture (ACL).
# with Eigen threadpool support
common:mkl_aarch64_threadpool --define=build_with_mkl_aarch64=true
common:mkl_aarch64_threadpool --@compute_library//:openmp=false
common:mkl_aarch64_threadpool -c opt

rank.fusion.window_size=200
# Rank constant for rank fusion.
rank.fusion.rank_constant=20
# Number of threads for rank fusion.
rank.fusion.threads=-1
# Score field for rank fusion.
rank.fusion.score_field=rf_score

# acl

# Whether to get SMB roles from a file.
smb.role.from.file=true
# Available SID types for SMB.
smb.available.sid.types=1,2,4:2,5:1
# Whether to get file roles from a file.
file.role.from.file=true

        back to a while loop.

*   XLA:

    *   `tf.distribute.MultiWorkerMirroredStrategy` is now compilable with XLA.
    *   [Compute Library for the Arm® Architecture (ACL)](https://github.com/ARM-software/ComputeLibrary)
        is supported for aarch64 CPU XLA runtime

*   CPU performance optimizations:

    *   **x86 CPUs**: