Skip to content

Instantly share code, notes, and snippets.

@sub-mod
sub-mod / tektoncd_notes.md
Last active January 29, 2020 20:16
tektoncd_notes.md
Starting local Bazel server and connecting to it...
{ "source_HEAD": "1c83755e5a1269b528042e6d318d66e363ffc349", "source_remote_origin": "https://github.com/tensorflow/tensorflow.git", "OS_VER": "CentOS release 6.10 (Final)", "GLIBC_VER": "ldd (GNU libc) 2.12", "PIP_VER": "pip 19.2.3 from /opt/rh/rh-python36/root/usr/lib/python3.6/site-packages/pip (python 3.6)", "PROTOC_VER": "libprotoc 2.3.0", "LOGICAL_CPUS": "64", "CORES_PER_PCPU": " 16", "PHYSICAL_CPUS": "2", "GCC_VER": "gcc (GCC) 8.2.1 20180905 (Red Hat 8.2.1-3)", "OS": "Linux", "kernel": "3.10.0-1058.el7.x86_64", "architecture": "broadwell", "processor": "Intel(R) Xeon(R) CPU E5-2697A v4 @ 2.60GHz", "Bazel_version": "Build label: 0.24.1- (@non-git)", "Java_version": "1.8.0_222", "Python_version": "3.6.3", "gpp_version": "g++ (GCC) 8.2.1 20180905 (Red Hat 8.2.1-3)", "swig_version": "", "NVIDIA_driver_version": "", "CUDA_device_count": "", "CUDA_device_names": "", "CUDA_toolkit_version": "V10.0.130", "GCC_FLAGS": "-march=broadwell -mmmx -msse -msse2 -mss
@sub-mod
sub-mod / tf_centos6_cuda.txt
Created September 17, 2019 03:56
TF centos6 cuda build fail
{ "source_HEAD": "1c83755e5a1269b528042e6d318d66e363ffc349", "source_remote_origin": "https://github.com/tensorflow/tensorflow.git", "OS_VER": "CentOS release 6.10 (Final)", "GLIBC_VER": "ldd (GNU libc) 2.12", "PIP_VER": "pip 19.2.3 from /opt/rh/rh-python36/root/usr/lib/python3.6/site-packages/pip (python 3.6)", "PROTOC_VER": "libprotoc 2.3.0", "LOGICAL_CPUS": "64", "CORES_PER_PCPU": " 16", "PHYSICAL_CPUS": "2", "GCC_VER": "gcc (GCC) 8.2.1 20180905 (Red Hat 8.2.1-3)", "OS": "Linux", "kernel": "3.10.0-1058.el7.x86_64", "architecture": "broadwell", "processor": "Intel(R) Xeon(R) CPU E5-2697A v4 @ 2.60GHz", "Bazel_version": "Build label: 0.24.1- (@non-git)", "Java_version": "1.8.0_222", "Python_version": "3.6.3", "gpp_version": "g++ (GCC) 8.2.1 20180905 (Red Hat 8.2.1-3)", "swig_version": "", "NVIDIA_driver_version": "", "CUDA_device_count": "", "CUDA_device_names": "", "CUDA_toolkit_version": "V10.0.130", "GCC_FLAGS": "-march=broadwell -mmmx -msse -msse2 -msse3 -mssse3 -mcx16 -msahf -mmovbe -maes -mpclmul -mpop
@sub-mod
sub-mod / tensorflow_txt.txt
Created September 16, 2019 15:24
tensorflow debugging
rm -fr /usr/bin/python
ln -s /usr/bin/python2.6 /usr/bin/python
export PYTHON_LIB_PATH=/usr/lib64/python2.6/site-packages
python -V
yum install devtoolset-8 prelink mlocate -y
export DEV_TOOLSET_VERSION=8
export JAVA_HOME=/usr/lib/jvm/java-1.8.0-openjdk-1.8.0.222.b10-0.el6_10.x86_64
export CUSTOM_BUILD='bazel build -c opt --config=cuda --cxxopt="-D_GLIBCXX_USE_CXX11_ABI=0" --verbose_failures //tensorflow/tools/pip_package:build_pip_package'
@sub-mod
sub-mod / crc_remote_laptop.md
Last active July 9, 2022 12:03
Setup CodeReady Containers on Remote Server and connect from Laptop 4.2

Following steps work for F29 for F31 look here

On the remote host

cat /etc/redhat-release
Fedora release 29 (Twenty Nine)

Install packages

docker run -it aicoe/tf-in-container:1.13.1-centos7 /bin/bash
[root@11174edc6cf7 /]#
[root@11174edc6cf7 /]# pip install ^C
[root@11174edc6cf7 /]# python -V
Python 2.7.5
[root@11174edc6cf7 /]# scl enable rh-python36
Need at least 3 arguments.
Run scl --help to get help.
[root@11174edc6cf7 /]# scl enable rh-python36 bash
[root@11174edc6cf7 /]# python -V
git checkout remotes/origin/v0.5-branch
cd components/tensorflow-notebook-image
./build_image.sh ./Dockerfile submod/tensorflow-1.13.1-notebook-cpu v0.5.0 ./versions/1.13.1/version-config.json
./build_image.sh ./Dockerfile submod/tensorflow-1.13.1-notebook-cpu v0.5.0 ./versions/1.13.1/version-config.json
+ DOCKERFILE=./Dockerfile
++ dirname ./Dockerfile
+ CONTEXT_DIR=.
+ IMAGE=submod/tensorflow-1.13.1-notebook-cpu
+ TAG=v0.5.0
+ CONFIG_FILE=./versions/1.13.1/version-config.json
tensorflow/core/kernels/unary_ops_composition.cc:432:1: required from here
tensorflow/core/kernels/unary_ops_composition.cc:140:20: warning: ignoring attributes on template argument 'Eigen::internal::packet_traits<float>::type {aka __vector(4) float}' [-Wignored-attributes]
INFO: From Compiling tensorflow/core/distributed_runtime/worker_session.cc [for host]:
In file included from ./tensorflow/core/platform/default/logging.h:24:0,
from ./tensorflow/core/platform/logging.h:25,
from ./tensorflow/core/framework/allocator.h:28,
from ./tensorflow/core/common_runtime/device.h:35,
from ./tensorflow/core/common_runtime/device_mgr.h:25,
from ./tensorflow/core/distributed_runtime/worker_session.h:21,
from tensorflow/core/distributed_runtime/worker_session.cc:15:
gcc version 7.3.1 20180303 (Red Hat 7.3.1-5) (GCC)
https://bugzilla.redhat.com/show_bug.cgi?id=1570308
external/eigen_archive/unsupported/Eigen/CXX11/src/Tensor/TensorDevice.h:35:62: required from 'Eigen::TensorDevice<ExpressionType, DeviceType>& Eigen::TensorDevice<ExpressionType, DeviceType>::operator=(const OtherDerived&) [with OtherDerived = Eigen::TensorMirrorPadOp<Eigen::array<Eigen::IndexPair<int>, 4>, const Eigen::TensorMap<Eigen::Tensor<const double, 4, 1, int>, 16, Eigen::MakePointer> >; ExpressionType = Eigen::TensorMap<Eigen::Tensor<double, 4, 1, int>, 16, Eigen::MakePointer>; DeviceType = Eigen::ThreadPoolDevice]'
./tensorflow/core/kernels/mirror_pad_op.h:351:27: required from 'void tensorflow::functor::MirrorPad<Device, T, Tpaddings, Dims>::operator()(const Device&, typename tensorflow::TTypes<T, Dims, int>::Tensor, typename tensorflow::TTypes<T, Dims, int>::ConstTensor, typename tensorflow::TTypes<Tpaddings>::ConstMatrix, int) [with Device = Eigen::ThreadPoolDevice; T = double; Tpaddin