The following changes are included:
- Starting from TF2.13v, TensorFlow uses Clang as compiler for Linux. Thus, we update the toolchain in release_cpu_linux and release_gpu_linux configs.
- Preserve the old linux build options in the unsupported_cpu_linux and unsupported_gpu_linux configs. If your project fails to build with Clang, you can use these unsupported flags to replace the release flags in your build command. However, please note that the old toolchain is no longer officially supported by TensorFlow and the unsupported configs will be removed soon. We strongly recommend that you migrate to Clang as your compiler for TensorFlow Linux builds. Instructions are available in the official documentation: https://www.tensorflow.org/install/source#install_clang_recommended_linux_only. Another good alternative is to use our Docker containers to build and test TensorFlow: https://github.com/tensorflow/tensorflow/tree/master/tensorflow/tools/tf_sig_build_dockerfiles.
- Add official linker options and container environment settings to release linux configs.
- Deduplicate build options in cpu.bazelrc and gpu.bazelrc.
- Delete outdated CI jobs.

PiperOrigin-RevId: 564620926

This marks an important step towards delivering OpenXLA, Google’s OSS-based unified ML software infrastructure.

PiperOrigin-RevId: 563177047

PiperOrigin-RevId: 561438023

TensorFlow no longer supports GPU on native-Windows since 2.11 (https://github.com/tensorflow/tensorflow/releases/tag/v2.11.1).

PiperOrigin-RevId: 561366344

startup options cannot be affected by --config=... flags, as they are initialized before the configs are evaluated.

PiperOrigin-RevId: 559586164

Changes include the following:

- Merge `rbe_linux_py3` and `rbe_linux_base` into `rbe_linux`
- Rename the `rbe` config defined in the root bazelrc to `rbe_base`. This avoids loops when the `rbe` config in cpu/cuda.bazelrc depends on rbe configs in the root bazelrc. There will be a followup change that completely deprecate `rbe` config name
- Rename `rbe_cpu_linux` to `rbe_linux_cpu` for consistency with the naming convention: `rbe_base` --> `rbe_linux` --> `rbe_linux_cpu` --> `rbe_linux_cuda`
- Enable TensorRT in the `cuda_clang` config
- Add `cuda_clang_official` to the root bazelrc to use CI toolchain and set compilation CUDA version and paths
- Add `rbe_linux_cuda` to the root bazelrc, which is a RBE option to build with GPU support using Clang
- Merge `tensorflow_testing_rbe_linux` into `rbe_linux_cpu` and `rbe_linux_cuda`
- Delete `rbe_linux_cuda11.2_nvcc*` and `rbe_linux_cuda_nvcc*` configs
- Update build commands in Kokoro jobs that are using the old rbe configs
- Deduplicate the rbe configs in the root, cpu, and cuda bazelrc files
- Update the toolchain and python settings in the root bazelrc, with clang as the default compiler and the python path to "/usr/bin/python3". I found a Kokoro job that failed to compile with clang, so I created a config `rbe_linux_cpu_old` using the old nvcc toolchain and python setup. This config will be removed when the compiler migration is complete.

PiperOrigin-RevId: 558863758

PiperOrigin-RevId: 553502182

PiperOrigin-RevId: 553201746

Change // to #

Temporarily reenable flaky_test_attempts for upcoming TensorFlow 2.14 release and branch cut window. This flag will be removed from master after branch cut.

PiperOrigin-RevId: 551548089

The following configs are removed:
- v1
- avx2_win
- avx2_linux
- native_arch_linux
- numa
- libc++
- ios_i386
- stackdriver_support
- rbe_lite_linux
- rbe_linux_cuda_nvcc
- rbe_gpu_linux
- rbe_linux_cuda11.2_nvcc_py3.8, rbe_linux_cuda_nvcc_py38
- rbe_linux_cuda11.2_nvcc_py3.10, rbe_linux_cuda_nvcc_py310
- rbe_linux_rocm_py3.7, rbe_linux_rocm_py3.8, rbe_linux_rocm_py3.10
- rbe_linux_cuda_clang_base, rbe_linux_cuda_clang_py**
- rbe_win_py37, rbe_win_py310

If the removal of a config breaks your workflow, you can add it back as a command line option. If you think the config is removed mistakenly, please open an issue on GitHub.

PiperOrigin-RevId: 549419745

PiperOrigin-RevId: 549134963

PiperOrigin-RevId: 545810433

"-c dbg" works for the xla repo, but not the tensorflow repo. Currently, --config=dbg works for tensorflow repo, but not the xla repo (--config=dbg is still defined in xla, but it doesn't work). It's hard to remember to use the right flag in the right repo. This PR makes --config=dbg work in the xla repo so that can be used in both tensorflow and xla.

Fixes https://github.com/openxla/xla/issues/3381.

PiperOrigin-RevId: 544435135

PiperOrigin-RevId: 543570166

PiperOrigin-RevId: 543555476

PiperOrigin-RevId: 540441442

PiperOrigin-RevId: 540321339

PiperOrigin-RevId: 539741752

Fixes https://github.com/tensorflow/tensorflow/issues/60508

PiperOrigin-RevId: 538899526

PiperOrigin-RevId: 538288491

pybind11 wrapper implementation for AotOptions,AotCompileSavedModel, and GraphExecutionOptions to ensure proper functionality.

PiperOrigin-RevId: 538193570

PiperOrigin-RevId: 535635561

PiperOrigin-RevId: 533555051

PiperOrigin-RevId: 532340336

PiperOrigin-RevId: 530484947

PiperOrigin-RevId: 529277193

PiperOrigin-RevId: 527625630

PiperOrigin-RevId: 525847365

PiperOrigin-RevId: 525598497

PiperOrigin-RevId: 525489441

PiperOrigin-RevId: 522403568

PiperOrigin-RevId: 520116023

Part of https://github.com/bazelbuild/bazel/issues/17731.

To build with oneDNN v3.x, pass --config=mkl --define=build_with_onednn_v3=true --define=build_with_onednn_v2=false

PiperOrigin-RevId: 512196114