Signed-off-by: Hongxu Jia <hongxu.jia@windriver.com>

Replace qemuarm override to arm, apply the fix to all 32 bit arm BSPs
Signed-off-by: Hongxu Jia <hongxu.jia@windriver.com>

Drop the revert patch, and fix the error directly

|tensorflow/compiler/mlir/quantization/tensorflow/debugging/mlir_dump.cc:93:10:
error: could not convert 'dump_file' from 'std::unique_ptr<llvm::raw_fd_ostream>'
to 'absl::lts_20220623::StatusOr<std::unique_ptr<llvm::raw_fd_ostream> >'
|   return dump_file;
          ^~~~~~~~~
Signed-off-by: Hongxu Jia <hongxu.jia@windriver.com>

Signed-off-by: Hongxu Jia <hongxu.jia@windriver.com>

Signed-off-by: Hongxu Jia <hongxu.jia@windriver.com>

The official tensorflow 2.12.0 is released [1], upgrade it

[1] https://github.com/tensorflow/tensorflow/releases

Signed-off-by: Hongxu Jia <hongxu.jia@windriver.com>

The recipe tensorflow does not support qemuarm, so drop
0001-fix-XNNPACK-build-failure-for-qemuarm.patch
Signed-off-by: Hongxu Jia <hongxu.jia@windriver.com>

$ find packages-split/ -name libtensorflow_cc.so*
packages-split/tensorflow-dbg/usr/lib64/.debug/libtensorflow_cc.so
packages-split/tensorflow-dbg/usr/lib64/python3.11/site-packages/tensorflow/.debug/libtensorflow_cc.so.2
packages-split/libtensorflow-c/usr/lib64/libtensorflow_cc.so
packages-split/python3-tensorflow/usr/lib64/python3.11/site-packages/tensorflow/libtensorflow_cc.so.2

Package libtensorflow-c provides /usr/lib64/libtensorflow_cc.so
and package python3-tensorflow provides /usr/lib64/python3.11/
site-packages/tensorflow/libtensorflow_cc.so.2

The OE build system's shared library resolver incorrectly report the
error of multiple shlib providers

|ERROR: python3-tensorflow: Multiple shlib providers for libtensorflow_cc.so.2:
libtensorflow-c, python3-tensorflow (used by files: tmp-glibc/work/
corei7-64-wrs-linux/tensorflow/2.12.0rc1-r0/packages-split/python3-tensorflow/
usr/lib64/python3.11/site-packages/tensorflow/python/_pywrap_tfe.so)

Ignore OE shared library resolver
Signed-off-by: Hongxu Jia <hongxu.jia@eng.windriver.com>

It does not make sense to run tensorflow on 32 bit arm bsp.
Signed-off-by: Hongxu Jia <hongxu.jia@windriver.com>

Signed-off-by: Hongxu Jia <hongxu.jia@windriver.com>

According to [1][2], adjust default ram and cpu for bazel,
use 25% of them for bazel build

[1] https://bazel.build/reference/command-line-reference#flag--local_ram_resources
[2] https://bazel.build/reference/command-line-reference#flag--local_cpu_resources

Signed-off-by: Hongxu Jia <hongxu.jia@eng.windriver.com>

For arm64 bsp, run 'bitbake lib32-tensorflow lib32-tensorflow-lite' caused
mutilple compile failures, it does not make sense to build lib32-tensorflow,
make recipe tensorflow not be built for multilib
Signed-off-by: Hongxu Jia <hongxu.jia@eng.windriver.com>

In commit [8235508f

 tensorflow: split sub packages], it split sub packages
python3-tensorflow out of package tensorflow.

This commit corrects the python3 runtime depends of packages
python3-tensorflow
Signed-off-by: Hongxu Jia <hongxu.jia@eng.windriver.com>

The tensorflow 2.12 has supported python 3.11, remove old python3
Signed-off-by: Hongxu Jia <hongxu.jia@eng.windriver.com>

Signed-off-by: Hongxu Jia <hongxu.jia@eng.windriver.com>

Signed-off-by: Hongxu Jia <hongxu.jia@eng.windriver.com>

Drop obsolete patch:
- 0001-support-to-compat-python-3.11.patch
Signed-off-by: Hongxu Jia <hongxu.jia@eng.windriver.com>

Remove python 3.10 requirement checking, tensorflow 2.12 has supported
python 3.11, remove old python 3.10

Refresh patches:
- 0001-add-yocto-toolchain-to-support-cross-compiling.patch
- 0001-support-32-bit-x64-and-arm-for-yocto.patch

Drop obsolete patches:
- 0003-Support-python-3.11-changes.patch
- 0004-protobuf-fix-build-with-Python-3.11.patch
- 0005-cpython-support-Python-3.11.patch
- 0006-wrapt-support-Python-3.11.patch

Add -Wno-stringop-overflow to CFLAGS to workaround gcc error
Signed-off-by: Hongxu Jia <hongxu.jia@eng.windriver.com>

Backport patches from python 3.11 to add a new standard library module,
`tomllib`, for parsing TOML.

The implementation is based on Tomli (https://github.com/hukkin/tomli

).
Signed-off-by: Hongxu Jia <hongxu.jia@eng.windriver.com>

Improve 0001-HttpDownloader-save-download-tarball-to-distdir.patch
to fix download failure of tensorflow 2.11.0

Issue: LINUXEXEC-24074
(LOCAL REV: NOT UPSTREAM) -- will send to upstream
Signed-off-by: Hongxu Jia <hongxu.jia@eng.windriver.com>

In tensorflow build, it downloads a full list of files, and
Bazel won't cache the downloaded archive.

But for Yocto, in order to support offline build ,we need to cache
downloaded archive, so hardcode go SDKS list files rather than
fetch from internet

Issue: LINUXEXEC-24074
(LOCAL REV: NOT UPSTREAM) -- will send to upstream
Signed-off-by: Hongxu Jia <hongxu.jia@eng.windriver.com>

Issue: LINUXEXEC-24074
(LOCAL REV: NOT UPSTREAM) -- will send to upstream
Signed-off-by: Hongxu Jia <hongxu.jia@eng.windriver.com>

Issue: LINUXEXEC-24074
(LOCAL REV: NOT UPSTREAM) -- will send to upstream
Signed-off-by: Hongxu Jia <hongxu.jia@eng.windriver.com>

Issue: LINUXEXEC-24074
(LOCAL REV: NOT UPSTREAM) -- will send to upstream
Signed-off-by: Hongxu Jia <hongxu.jia@eng.windriver.com>

Drop two backported patches:
- 0001-Fix-Python-3.11-compatibility-problems-for-JAX-XLA.patch
- 0002-Update-pybind11-to-2.10.0.patch

Refresh two patches:
- 0001-support-musl.patch
- 0001-support-32-bit-x64-and-arm-for-yocto.patch

Issue: LINUXEXEC-24074
(LOCAL REV: NOT UPSTREAM) -- will send to upstream
Signed-off-by: Hongxu Jia <hongxu.jia@eng.windriver.com>

Other than print mess of warning message during bitbake parsing,
directly skip tensorflow recipe if python3.10 is not used, in this
situation, there will be no warning message if tensorflow is not
explicitly built

$ bitbake tensorflow
ERROR: Nothing PROVIDES 'tensorflow'
tensorflow was skipped: We shouldn't have multilib variants for tensorflow
tensorflow was skipped:
The official TensorFlow is tested and supported under Python 3.7-3.10
Please add the following to local.conf
    DISTRO_FEATURES:append = ' tensorflow'
    DISTRO_FEATURES_NATIVE:append = ' tensorflow'
    DISTRO_FEATURES_NATIVESDK:append = ' tensorflow'
It will apply python3 3.10.6 recipe
Signed-off-by: Hongxu Jia <hongxu.jia@eng.windriver.com>

Add a checking tensorflow recipe to warn when tensorflow not in
DISTRO_FEATURES and PYTHON_BASEVERSION version is not 3.10

...
WARNING:meta-tensorflow/recipes-framework/tensorflow/tensorflow-lite_2.10.0.bb:
The official TensorFlow is tested and supported under Python 3.7-3.10
Please add DISTRO_FEATURES += 'tensorflow' to local.conf
It will apply python3 3.10.6 recipe
WARNING:meta-tensorflow/recipes-framework/tensorflow/tensorflow-lite_2.10.0.bb:
The official TensorFlow is tested and supported under Python 3.7-3.10
Please add the following to local.conf
    PYTHON_BASEVERSION:class-target = '3.10'
    PYTHON_BASEVERSION:class-native = '3.10'
    PYTHON_BASEVERSION:class-nativesdk = '3.10'
It will apply python3 modules for 3.10
...
Signed-off-by: Hongxu Jia <hongxu.jia@eng.windriver.com>

The official TensorFlow is tested and supported Python 3.7-3.10 [1][2]
Add python3 3.10 back for tensorflow

In order to distinguish with python3 3.11 in oe-core, add required distro
features `tensorflow' to python3 3.10.6 recipe.

Set the following in local.conf to apply python3 3.10.6

  DISTRO_FEATURES += "tensorflow"
  PYTHON_BASEVERSION:class-target = "3.10"
  PYTHON_BASEVERSION:class-native = "3.10"
  PYTHON_BASEVERSION:class-nativesdk = "3.10"

[1] https://www.tensorflow.org/install
[2] https://www.tensorflow.org/install/source#tested_build_configurations

Signed-off-by: Hongxu Jia <hongxu.jia@windriver.com>
Signed-off-by: Hongxu Jia <hongxu.jia@eng.windriver.com>

The SafeConfigParser class will be renamed to ConfigParser in Python
3.12 [1]. This alias will be removed in future versions.So we can use
ConfigParser directly instead.

[1] https://github.com/python/cpython/issues/89336

Signed-off-by: Hongxu Jia <hongxu.jia@windriver.com>

The python 3.11 has removed long-deprecated inspect methods [1],
use collections.namedtuple to instead

[1] https://github.com/python/cpython/commit/d89fb9a5a610a257014d112bdceef73d7df14082

Signed-off-by: Hongxu Jia <hongxu.jia@windriver.com>

After python3 in Yocto was upgraded to 3.11, the result of
`importlib.import_module('distutils.sysconfig').get_python_inc()'
does not right, but the result of
`importlib.import_module('sysconfig').get_path('include')'
is OK

$ ../recipe-sysroot-native/usr/bin/python3-native/python3
>>> import importlib; import importlib.util;
>>> importlib.import_module('distutils.sysconfig').get_python_inc()
'/tmp-glibc/work/x86_64-linux/tensorflow-native/2.10.0-r0/recipe-sysroot-native/tmp-glibc/work/x86_64-linux/tensorflow-native/2.10.0-r0/recipe-sysroot-native/usr/include/python3.11'
>>> importlib.import_module('sysconfig').get_path('include')
'/tmp-glibc/work/x86_64-linux/tensorflow-native/2.10.0-r0/recipe-sysroot-native/usr/include/python3.11'
Signed-off-by: Hongxu Jia <hongxu.jia@windriver.com>

Backport patches from upstream to fix Python 3.11 compatibility problems
Signed-off-by: Hongxu Jia <hongxu.jia@windriver.com>

...
|external/XNNPACK/src/qs8-igemm/gen/4x8c4-minmax-rndnu-aarch32-neondot-ld64.S:140:
Error: selected processor does not support `vsdot.s8 q8,q2,d0[0]' in ARM
mode
...

Does not support 32 bit ARM with Neon DotProduct
Signed-off-by: Hongxu Jia <hongxu.jia@windriver.com>

...
external/XNNPACK/src/f32-f16-vcvt/gen/vcvt-neonfp16-x16.c:36:28: error: incompatible types when initializing type 'uint16x8_t' using type 'int'
   36 |     const uint16x8_t vh0 = vreinterpretq_u16_f16(vcombine_f16(vcvt_f16_f32(vf0), vcvt_f16_f32(vf1)));
      |                            ^~~~~~~~~~~~~~~~~~~~~
...

See [1][2] for details

[1] https://github.com/tensorflow/tensorflow/issues/54268
[2] https://github.com/tensorflow/tensorflow/pull/54316/commits/abae5a1e556b4e595a84e9a25fb1e9e86d7cbd6b

Signed-off-by: Hongxu Jia <hongxu.jia@windriver.com>

Signed-off-by: Kai Kang <kai.kang@windriver.com>

Refresh patches
- 0001-drop-bazel_toolchains_repositories-to-support-offlin.patch
- 0001-support-32-bit-x64-and-arm-for-yocto.patch
- 0001-third_party-eigen_archive-workaround-ice-failure-whi.patch

Add python3-requests-native to depends
Signed-off-by: Hongxu Jia <hongxu.jia@windriver.com>

Signed-off-by: Hongxu Jia <hongxu.jia@windriver.com>

Signed-off-by: Hongxu Jia <hongxu.jia@windriver.com>

Drop obsolete patch
- 0001-update-link-of-zlib-1.2.11.tar.gz.patch
Signed-off-by: Hongxu Jia <hongxu.jia@windriver.com>

It separated python module from flatbuffers and created a new recipe
python3-flatbuffers in meta-openembedded. Update dependency of keras
accordingly.
Signed-off-by: Kai Kang <kai.kang@windriver.com>