Warning while creating Session on Mac OS X: can't determine number of CPU cores #27

delip · 2015-11-09T18:55:02Z

While creating a Session, I get this warning (using pre-built whl on OSX). Any ideas?

s = tf.Session()
can't determine number of CPU cores: assuming 4
I tensorflow/core/common_runtime/local_device.cc:25] Local device intra op parallelism threads: 4
can't determine number of CPU cores: assuming 4
I tensorflow/core/common_runtime/local_session.cc:45] Local session inter op parallelism threads: 4

The text was updated successfully, but these errors were encountered:

vrv · 2015-11-09T18:58:07Z

Hi delip, we fixed this in 430a054 -- when we mint our next version of the pip binaries, this should be fixed. Thanks for your report!

delip · 2015-11-09T23:17:47Z

Excellent!

gynsolomon · 2015-11-10T05:16:42Z

thanks!

delip · 2015-11-17T18:15:14Z

Just want to confirm that this warning went away with the latest build. I will be closing this. Thanks!

…'s bug there though)

…ase of LIBXSMM_DNN_WARN_FALLBACK, use libxsmm_hash instead of std::hash, code cleanup (tensorflow#27) * Fixed AVX-512 intrinsic implementation. * OR'ed LIBXSMM_DNN_CONV_OPTION_OVERWRITE into convolution options, which folds zeroing the input buffer on first use. This removes the call to libxsmm_dnn_zero_buffer in case of LIBXSMM_DNN_COMPUTE_KIND_FWD. * Rely on libxsmm_hash rather than std::hash. Brought xsmm_conv2d.cc up-to-date with TF/master. * Code cleanup: use LIBXSMM_DNN_CONV_OPTION_WU_EXT_FILTER_REDUCE_OVERWRITE rather than assembling the option from separate flags. * Avoid to destroy the handle in case of LIBXSMM_DNN_WARN_FALLBACK since the next iteration may double-delete the same handle. One would need to update the handle-cache to allow destruction at this place. However, all handles are destructed when TF terminates (cache cleanup).

* Fixed AVX-512 intrinsic layer (sparse_matmul_op.h). Incorporated LIBXSMM_DNN_CONV_OPTION_OVERWRITE. (#26) * Fixed AVX-512 intrinsic implementation. * OR'ed LIBXSMM_DNN_CONV_OPTION_OVERWRITE into convolution options, which folds zeroing the input buffer on first use. This removes the call to libxsmm_dnn_zero_buffer in case of LIBXSMM_DNN_COMPUTE_KIND_FWD. * Made xsmm_conv2d.cc up-to-date with TF/master, avoid double-free in case of LIBXSMM_DNN_WARN_FALLBACK, use libxsmm_hash instead of std::hash, code cleanup (#27) * Fixed AVX-512 intrinsic implementation. * OR'ed LIBXSMM_DNN_CONV_OPTION_OVERWRITE into convolution options, which folds zeroing the input buffer on first use. This removes the call to libxsmm_dnn_zero_buffer in case of LIBXSMM_DNN_COMPUTE_KIND_FWD. * Rely on libxsmm_hash rather than std::hash. Brought xsmm_conv2d.cc up-to-date with TF/master. * Code cleanup: use LIBXSMM_DNN_CONV_OPTION_WU_EXT_FILTER_REDUCE_OVERWRITE rather than assembling the option from separate flags. * Avoid to destroy the handle in case of LIBXSMM_DNN_WARN_FALLBACK since the next iteration may double-delete the same handle. One would need to update the handle-cache to allow destruction at this place. However, all handles are destructed when TF terminates (cache cleanup). * Configure LIBXSMM with default arguments (#28) * Fixed AVX-512 intrinsic implementation. * OR'ed LIBXSMM_DNN_CONV_OPTION_OVERWRITE into convolution options, which folds zeroing the input buffer on first use. This removes the call to libxsmm_dnn_zero_buffer in case of LIBXSMM_DNN_COMPUTE_KIND_FWD. * Rely on libxsmm_hash rather than std::hash. Brought xsmm_conv2d.cc up-to-date with TF/master. * Code cleanup: use LIBXSMM_DNN_CONV_OPTION_WU_EXT_FILTER_REDUCE_OVERWRITE rather than assembling the option from separate flags. * Avoid to destroy the handle in case of LIBXSMM_DNN_WARN_FALLBACK since the next iteration may double-delete the same handle. One would need to update the handle-cache to allow destruction at this place. However, all handles are destructed when TF terminates (cache cleanup). * Rely on default configuration arguments, and thereby lower the dependence from LIBXSMM internals.

…ase of LIBXSMM_DNN_WARN_FALLBACK, use libxsmm_hash instead of std::hash, code cleanup (tensorflow#27) * Fixed AVX-512 intrinsic implementation. * OR'ed LIBXSMM_DNN_CONV_OPTION_OVERWRITE into convolution options, which folds zeroing the input buffer on first use. This removes the call to libxsmm_dnn_zero_buffer in case of LIBXSMM_DNN_COMPUTE_KIND_FWD. * Rely on libxsmm_hash rather than std::hash. Brought xsmm_conv2d.cc up-to-date with TF/master. * Code cleanup: use LIBXSMM_DNN_CONV_OPTION_WU_EXT_FILTER_REDUCE_OVERWRITE rather than assembling the option from separate flags. * Avoid to destroy the handle in case of LIBXSMM_DNN_WARN_FALLBACK since the next iteration may double-delete the same handle. One would need to update the handle-cache to allow destruction at this place. However, all handles are destructed when TF terminates (cache cleanup).

Fix typo mistake, fixes tensorflow#27.

* [OpenCL] Registers ApplyMomentum * Deleting code * Fixed errors (still runs on CPU) * Changed implementation of ApplyMomentum

Update tensorflow-install-basic.md

…ase of LIBXSMM_DNN_WARN_FALLBACK, use libxsmm_hash instead of std::hash, code cleanup (tensorflow#27) * Fixed AVX-512 intrinsic implementation. * OR'ed LIBXSMM_DNN_CONV_OPTION_OVERWRITE into convolution options, which folds zeroing the input buffer on first use. This removes the call to libxsmm_dnn_zero_buffer in case of LIBXSMM_DNN_COMPUTE_KIND_FWD. * Rely on libxsmm_hash rather than std::hash. Brought xsmm_conv2d.cc up-to-date with TF/master. * Code cleanup: use LIBXSMM_DNN_CONV_OPTION_WU_EXT_FILTER_REDUCE_OVERWRITE rather than assembling the option from separate flags. * Avoid to destroy the handle in case of LIBXSMM_DNN_WARN_FALLBACK since the next iteration may double-delete the same handle. One would need to update the handle-cache to allow destruction at this place. However, all handles are destructed when TF terminates (cache cleanup).

Add Boost.Locale as a dependency for the build

vrv changed the title ~~Warning while creating Session~~ Warning while creating Session on Mac OS X: can't determine number of CPU cores Nov 9, 2015

delip closed this as completed Nov 17, 2015

ilblackdragon added a commit to ilblackdragon/tensorflow that referenced this issue Mar 9, 2016

Fixes tensorflow#27: Added alternative import for v0.6 TensorFlow (it…

da88e3f

…'s bug there though)

wangyongliang mentioned this issue Apr 27, 2016

core dump when import tensorflow #2129

Closed

bluezone2015 mentioned this issue Mar 8, 2017

import tensorflow Segmentation fault (core dumped) #8197

Closed

hexujun mentioned this issue Apr 10, 2017

android libtensorflow_inference.so run crash (signal 6 (SIGABRT)) #9096

Closed

tarasglek pushed a commit to tarasglek/tensorflow that referenced this issue Jun 20, 2017

Merge pull request tensorflow#28 from LiberiFatali/master

1d1b340

Fix typo mistake, fixes tensorflow#27.

zhangdingfei mentioned this issue Jun 21, 2017

tensorflow-1.2.0 import tensorflow Segmentation fault #10870

Closed

jakiechris mentioned this issue Sep 4, 2017

protobuf crashes at runtime when loading tensor lib. #12794

Closed

lukeiwanski referenced this issue in codeplaysoftware/tensorflow Oct 26, 2017

[OpenCL] Registers ApplyMomentum (#27)

f423f2c

* [OpenCL] Registers ApplyMomentum * Deleting code * Fixed errors (still runs on CPU) * Changed implementation of ApplyMomentum

whchung referenced this issue in ROCm/tensorflow-upstream Apr 10, 2018

Merge pull request #27 from ROCmSoftwarePlatform/parallelo-patch-2

eaef8ea

Update tensorflow-install-basic.md

ychen404 mentioned this issue Aug 19, 2018

Not able to port a 6-layered mobilenet tflite model to mobile #21368

Closed

chenjiasheng mentioned this issue Dec 12, 2018

Distributed Training Randomly Stops During the Training Process #12667

Closed

eggonlea pushed a commit to eggonlea/tensorflow that referenced this issue Mar 12, 2019

Merge pull request tensorflow#27 from mozilla/boost-dep

0b39249

Add Boost.Locale as a dependency for the build

isra60 mentioned this issue Mar 25, 2019

Segmentation Fault with TensorRT create interference graph #27100

Closed

dkashkin mentioned this issue Apr 25, 2019

TFLite Interpreter fails to load quantized model on Android (stock ssd_mobilenet_v2) #28163

Closed

chengdianxuezi mentioned this issue Nov 1, 2019

Bug: tensorflow-gpu takes long time before beginning to compute #18652

Closed

yanceyblog mentioned this issue Nov 28, 2019

armeabi-v7a libtensorflowlite_jni.so：signal 7 (SIGBUS), code 1 (BUS_ADRALN), fault addr 0xeef5445f #34669

Closed

dinkdeep mentioned this issue Apr 7, 2021

Segmentation fault in tf-opt while running a tf dialect mlir file #48365

Open

rsanthanam-amd mentioned this issue Jul 1, 2021

[ROCm] This change replaces the original assert for detecting multiple #49232

Closed

ivankxt mentioned this issue Jun 12, 2023

Get deadlock after Predict(cuda10.0, cudnn7.6.5, Tesla T4 GPU) #60841

Closed

lyz1005 mentioned this issue Oct 26, 2023

Interpreter run crash #62240

Closed

spacycoder mentioned this issue Dec 11, 2023

Why does my full integer quantized tflite model crash when loaded? #62618

Closed

ramonhollands mentioned this issue Jun 30, 2024

Tflite model fails on Android GPU delegate due to 'null pointer dereference' #70664

Open

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Warning while creating Session on Mac OS X: can't determine number of CPU cores #27

Warning while creating Session on Mac OS X: can't determine number of CPU cores #27

delip commented Nov 9, 2015

vrv commented Nov 9, 2015

delip commented Nov 9, 2015

gynsolomon commented Nov 10, 2015

delip commented Nov 17, 2015

Warning while creating Session on Mac OS X: can't determine number of CPU cores #27

Warning while creating Session on Mac OS X: can't determine number of CPU cores #27

Comments

delip commented Nov 9, 2015

vrv commented Nov 9, 2015

delip commented Nov 9, 2015

gynsolomon commented Nov 10, 2015

delip commented Nov 17, 2015