private final Range<C> restriction;

    SubRangeSet(Range<C> restriction) {
      super(
          new SubRangeSetRangesByLowerBound<C>(
              Range.<Cut<C>>all(), restriction, TreeRangeSet.this.rangesByLowerBound));
      this.restriction = restriction;
    }

    @Override
    public boolean encloses(Range<C> range) {
      if (!restriction.isEmpty() && restriction.encloses(range)) {

    /** Logon failure: unknown user name or bad password */
    int NT_STATUS_LOGON_FAILURE = 0xC000006d;
    /** Logon failure: user account restriction */
    int NT_STATUS_ACCOUNT_RESTRICTION = 0xC000006e;
    /** Logon failure: account logon time restriction violation */
    int NT_STATUS_INVALID_LOGON_HOURS = 0xC000006f;
    /** Logon failure: user not allowed to log on to this computer */

    /** Logon failure: unknown user name or bad password */
    int NT_STATUS_LOGON_FAILURE = 0xC000006d;
    /** Logon failure: user account restriction */
    int NT_STATUS_ACCOUNT_RESTRICTION = 0xC000006e;
    /** Logon failure: account logon time restriction violation */
    int NT_STATUS_INVALID_LOGON_HOURS = 0xC000006f;
    /** Logon failure: user not allowed to log on to this computer */

)

# Initialize the TensorFlow repository and all dependencies.
#
# The cascade of load() statements and tf_workspace?() calls works around the
# restriction that load() statements need to be at the top of .bzl files.
# E.g. we can not retrieve a new repository with http_archive and then load()
# a macro from that repository in the same file.

#
# Macosx options
#     darwin_arm64:
#
# Compiler options:
#     cuda_clang:             Use Clang when building CUDA code.
#     avx_linux:              Build with avx instruction set on linux.
#     avx_win:                Build with avx instruction set on windows
#
# Other build options:
#     short_logs:       Only log errors during build, skip warnings.
#     verbose_logs:     Show all compiler warnings during build.

///

/// warning

Make sure to **always** use the **exec form** of the `CMD` instruction, as explained below.

///

#### Use `CMD` - Exec Form { #use-cmd-exec-form }

The <a href="https://docs.docker.com/reference/dockerfile/#cmd" class="external-link" target="_blank">`CMD`</a> Docker instruction can be written using two forms:

✅ **Exec** form:

```Dockerfile
# ✅ Do this

	MOVWP	(R5), R4		// a4000024
	MOVVP	-32768(R5), R4		// a4008026
	MOVVP	32764(R5), R4		// a4fc7f26
	MOVVP	32(R5), R4		// a4200026
	MOVVP	4(R5), R4		// a4040026
	MOVVP	(R5), R4		// a4000026

	// ADDU16I.D instruction
	ADDV16	$(-32768<<16), R4, R5	// ADDV16	$-2147483648, R4, R5	// 85000012
	ADDV16	$(0<<16), R4, R5	// ADDV16	$0, R4, R5		// 85000010
	ADDV16	$(8<<16), R4, R5	// ADDV16	$524288, R4, R5		// 85200010

	if srcInfo.Legacy {
		srcInfo.metadataOnly = false
	}

	// Check if x-amz-metadata-directive or x-amz-tagging-directive was not set to REPLACE and source,
	// destination are same objects. Apply this restriction also when
	// metadataOnly is true indicating that we are not overwriting the object.
	// if encryption is enabled we do not need explicit "REPLACE" metadata to
	// be enabled as well - this is to allow for key-rotation.

- Allowed dynamic configuration of the service account name and audience that the kubelet could request a token for, as part of the node audience restriction feature. ([#130485](https://github.com/kubernetes/kubernetes/pull/130485), [@aramase](https://github.com/aramase)) [SIG Auth and Testing]

   available for easier TPU based installs.
*  TensorFlow pip packages are now built with CUDA 12.3 and cuDNN 8.9.7
*  Added experimental support for float16 auto-mixed precision using the new AMX-FP16 instruction set on X86 CPUs.


### Bug Fixes and Other Changes

* `tf.lite`
    * Added support for `stablehlo.gather`.
    * Added support for `stablehlo.add`.
    * Added support for `stablehlo.multiply`.