Can't install from source if I don't have a GPU? #71

spk921 · 2015-11-10T03:47:10Z

My os X: EI Captian version 10.11.1
I tried install from the source. However, running "$ bazel-bin/tensorflow/tools/pip_package/build_pip_package /tmp/tensorflow_pkg" this make me error. I got following error message. I think I don't have a GPU in my mac. Is there anyway that I can isnstall from source even though I don't have a GPU ?

sangpil_kim@ ~/git/tensorflow $ bazel build -c opt //tensorflow/tools/pip_package:build_pip_package
.......
INFO: Found 1 target...
INFO: From Executing genrule //third_party/gpus/cuda:cuda_config_check:
/bin/bash: touch: command not found
ERROR: /Users/sangpil_kim/git/tensorflow/third_party/gpus/cuda/BUILD:143:1: Executing genrule //third_party/gpus/cuda:cuda_config_check failed: bash failed: error executing command /bin/bash -c 'source tools/genrule/genrule-setup.sh; touch bazel-out/local_darwin-opt/genfiles/third_party/gpus/cuda/cuda.config': com.google.devtools.build.lib.shell.BadExitStatusException: Process exited with status 127.
Target //tensorflow/tools/pip_package:build_pip_package failed to build
Use --verbose_failures to see the command lines of failed build steps.
INFO: Elapsed time: 79.345s, Critical Path: 0.32s
-bash: tail: command not found
-bash: sed: command not found

vincentvanhoucke · 2015-11-10T15:03:18Z

I suspect the GPU is a red herring. Without more info, the fact that /bin/bash can't seem to be able to execute standard tools like 'touch', 'tail' or 'sed' suggests something is very weird about your system.

booherbg · 2015-11-10T18:58:27Z

The actual error is that the 'touch' command fails. Fix that and you should be ok?

spk921 · 2015-11-11T12:38:36Z

Thanks, after that error my bash cracked so I just reinstalled my OS. I hope to know that there is a case who build from source which don't have a GPU in their labtop.

mrry · 2015-11-12T18:58:11Z

After reinstalling your OS, are you able to run the commands /usr/bin/touch, /usr/bin/tail and /usr/bin/sed from the Terminal? If so, this step in the installation should succeed.

To clarify: it is possible to build TensorFlow from source without a GPU.

spk921 · 2015-11-13T12:13:21Z

I used Time Machine before reinstalling it and I found those commands works. I assume my errors are related with CUDA configuration. I don't have a GPU but I installed CUDA so TensorFlow act like I have a GPU and makes a problem. However, I know it's possible to build without GPU. I will try to solve it next time when I have a time. I saw a benchmark that TensorFlow is not fast as torch so I decided to stick with Torch. I will try to install TensorFlow next time when it get better in the future.

This commit improves JitRunner so that it creates a target machine for the current CPU host which is used to properly initialize LLVM's TargetTransformInfo for such a target. This will enable optimizations such as vectorization in LLVM when using JitRunner. Please, note that, as part of this work, JITTargetMachineBuilder::detectHost() has been extended to include the host CPU name and sub-target features as part of the host CPU detection (https://reviews.llvm.org/D65760). Closes #71 COPYBARA_INTEGRATE_REVIEW=tensorflow/mlir#71 from dcaballe:dcaballe/tm_runner fadb4af5f0c78c0dd0d897395af76809f67f5bab PiperOrigin-RevId: 262452525

…lgorithmconfig-profileresult Refactor AlgorithmConfig / ProfileResult and put scratch size in AlgorithmDesc

PiperOrigin-RevId: 449112147

Imported from GitHub PR #74506 - Adds bf16, f16 support for tfl.atan2 - Adds bf16,f16,int8,int16 for tfl.neg - Adds bf16, f16 support for tfl.min, tfl.max - Adds bf16, f16 support for tfl.slice - Adds bf16, f16 support for tfl.round - Adds bf16, f16 support for tfl.reverse - Adds bf16, f16 support for tfl.pad - Adds bf16, f16 support for tfl.tanh and tfl.logistic - Adds bf16, f16 support for tfl.floor Copybara import of the project: -- 775ae2b by swatheesh-mcw <swatheesh.muralidharan@multicorewareinc.com>: TfLite neg_op add type support (#77) * TfLite neg_op add type support -Adds type support for bfloat16 -Adds type supoort for float16 -Adds type support for Int8 -Adds type support for Int16 Co-authored-by: pratham-mcw <pratham.kumar@multicorewareinc.com> -- e108d46 by swatheesh-mcw <swatheesh.muralidharan@multicorewareinc.com>: Tflite tanh and logistic adds missing datatype support (#76) * Tflite tanh and logistic adds missing datatype support -Adds f16,bf16 for tanh and logistic -Adds f16,bf16 for tanh and logistic unit test Co-authored-by: nitheshsrikanth-mcw <nithesh.srikanth@multicorewareinc.com> -- 18581a6 by swatheesh-mcw <swatheesh.muralidharan@multicorewareinc.com>: Tflite floor adds missing datatype support (#75) * Tflite floor missing datatype support -Adds f16,bf16 for floor -Adds f16,bf16 for floor unit test Co-authored-by: nitheshsrikanth-mcw <nithesh.srikanth@multicorewareinc.com> -- e0aaea1 by swatheesh-mcw <swatheesh.muralidharan@multicorewareinc.com>: TfLite Slice missing datatype support (#33) (#74) * TfLite Slice missing datatype support -Adds bf16, f16 support for slice -Adds bf16, f16 slice unit tests -- 458c6a4 by swatheesh-mcw <swatheesh.muralidharan@multicorewareinc.com>: TfLite Round missing datatype support (#32) (#73) * TfLite Round missing datatype support -Adds bf16, f16 support for round -Adds bf16, f16 round unit tests -- 5c6def1 by swatheesh-mcw <swatheesh.muralidharan@multicorewareinc.com>: Tflite min max missing datatypes support (#38) (#72) -Adds bf16,f16 for tflite min max operations -Adds bf16,f16 min max unit tests -- baea4f1 by swatheesh-mcw <swatheesh.muralidharan@multicorewareinc.com>: TfLite Reverse missing datatype support (#31) (#71) * TfLite Reverse missing datatype support -Adds bf16, f16 support for reverse -Adds bf16, f16 reverse unit tests -- 96ccf49 by Swatheesh Muralidharan <swatheesh.muralidharan@multicorewareinc.com>: Addressed redefinition of TFLITE_TENSOR_TYPE_ASSOC error in test_util.h file -- 2ff70a8 by RahulSudarMCW <rahul.sundar@multicorewareinc.com>: Include non-quantized int8 & int16 type support for tfl.pad Merging this change closes #74506 FUTURE_COPYBARA_INTEGRATE_REVIEW=#74506 from MCW-Dev:master 2ff70a8 PiperOrigin-RevId: 733452310

Imported from GitHub PR #74506 - Adds bf16, f16 support for tfl.atan2 - Adds bf16,f16,int8,int16 for tfl.neg - Adds bf16, f16 support for tfl.min, tfl.max - Adds bf16, f16 support for tfl.slice - Adds bf16, f16 support for tfl.round - Adds bf16, f16 support for tfl.reverse - Adds bf16, f16 support for tfl.pad - Adds bf16, f16 support for tfl.tanh and tfl.logistic - Adds bf16, f16 support for tfl.floor Copybara import of the project: -- 775ae2b by 8000 swatheesh-mcw <swatheesh.muralidharan@multicorewareinc.com>: TfLite neg_op add type support (#77) * TfLite neg_op add type support -Adds type support for bfloat16 -Adds type supoort for float16 -Adds type support for Int8 -Adds type support for Int16 Co-authored-by: pratham-mcw <pratham.kumar@multicorewareinc.com> -- e108d46 by swatheesh-mcw <swatheesh.muralidharan@multicorewareinc.com>: Tflite tanh and logistic adds missing datatype support (#76) * Tflite tanh and logistic adds missing datatype support -Adds f16,bf16 for tanh and logistic -Adds f16,bf16 for tanh and logistic unit test Co-authored-by: nitheshsrikanth-mcw <nithesh.srikanth@multicorewareinc.com> -- 18581a6 by swatheesh-mcw <swatheesh.muralidharan@multicorewareinc.com>: Tflite floor adds missing datatype support (#75) * Tflite floor missing datatype support -Adds f16,bf16 for floor -Adds f16,bf16 for floor unit test Co-authored-by: nitheshsrikanth-mcw <nithesh.srikanth@multicorewareinc.com> -- e0aaea1 by swatheesh-mcw <swatheesh.muralidharan@multicorewareinc.com>: TfLite Slice missing datatype support (#33) (#74) * TfLite Slice missing datatype support -Adds bf16, f16 support for slice -Adds bf16, f16 slice unit tests -- 458c6a4 by swatheesh-mcw <swatheesh.muralidharan@multicorewareinc.com>: TfLite Round missing datatype support (#32) (#73) * TfLite Round missing datatype support -Adds bf16, f16 support for round -Adds bf16, f16 round unit tests -- 5c6def1 by swatheesh-mcw <swatheesh.muralidharan@multicorewareinc.com>: Tflite min max missing datatypes support (#38) (#72) -Adds bf16,f16 for tflite min max operations -Adds bf16,f16 min max unit tests -- baea4f1 by swatheesh-mcw <swatheesh.muralidharan@multicorewareinc.com>: TfLite Reverse missing datatype support (#31) (#71) * TfLite Reverse missing datatype support -Adds bf16, f16 support for reverse -Adds bf16, f16 reverse unit tests -- 96ccf49 by Swatheesh Muralidharan <swatheesh.muralidharan@multicorewareinc.com>: Addressed redefinition of TFLITE_TENSOR_TYPE_ASSOC error in test_util.h file -- 2ff70a8 by RahulSudarMCW <rahul.sundar@multicorewareinc.com>: Include non-quantized int8 & int16 type support for tfl.pad Merging this change closes #74506 FUTURE_COPYBARA_INTEGRATE_REVIEW=#74506 from MCW-Dev:master 2ff70a8 PiperOrigin-RevId: 750807771

Imported from GitHub PR #74506 - Adds bf16, f16 support for tfl.atan2 - Adds bf16,f16,int8,int16 for tfl.neg - Adds bf16, f16 support for tfl.min, tfl.max - Adds bf16, f16 support for tfl.slice - Adds bf16, f16 support for tfl.round - Adds bf16, f16 support for tfl.reverse - Adds bf16, f16 support for tfl.pad - Adds bf16, f16 support for tfl.tanh and tfl.logistic - Adds bf16, f16 support for tfl.floor Copybara import of the project: -- 775ae2b by swatheesh-mcw <swatheesh.muralidharan@multicorewareinc.com>: TfLite neg_op add type support (#77) * TfLite neg_op add type support -Adds type support for bfloat16 -Adds type supoort for float16 -Adds type support for Int8 -Adds type support for Int16 Co-authored-by: pratham-mcw <pratham.kumar@multicorewareinc.com> -- e108d46 by swatheesh-mcw <swatheesh.muralidharan@multicorewareinc.com>: Tflite tanh and logistic adds missing datatype support (#76) * Tflite tanh and logistic adds missing datatype support -Adds f16,bf16 for tanh and logistic -Adds f16,bf16 for tanh and logistic unit test Co-authored-by: nitheshsrikanth-mcw <nithesh.srikanth@multicorewareinc.com> -- 18581a6 by swatheesh-mcw <swatheesh.muralidharan@multicorewareinc.com>: Tflite floor adds missing datatype support (#75) * Tflite floor missing datatype support -Adds f16,bf16 for floor -Adds f16,bf16 for floor unit test Co-authored-by: nitheshsrikanth-mcw <nithesh.srikanth@multicorewareinc.com> -- e0aaea1 by swatheesh-mcw <swatheesh.muralidharan@multicorewareinc.com>: TfLite Slice missing datatype support (#33) (#74) * TfLite Slice missing datatype support -Adds bf16, f16 support for slice -Adds bf16, f16 slice unit tests -- 458c6a4 by swatheesh-mcw <swatheesh.muralidharan@multicorewareinc.com>: TfLite Round missing datatype support (#32) (#73) * TfLite Round missing datatype support -Adds bf16, f16 support for round -Adds bf16, f16 round unit tests -- 5c6def1 by swatheesh-mcw <swatheesh.muralidharan@multicorewareinc.com>: Tflite min max missing datatypes support (#38) (#72) -Adds bf16,f16 for tflite min max operations -Adds bf16,f16 min max unit tests -- baea4f1 by swatheesh-mcw <swatheesh.muralidharan@multicorewareinc.com>: TfLite Reverse missing datatype support (#31) (#71) * TfLite Reverse missing datatype support -Adds bf16, f16 support for reverse -Adds bf16, f16 reverse unit tests -- 96ccf49 by Swatheesh Muralidharan <swatheesh.muralidharan@multicorewareinc.com>: Addressed redefinition of TFLITE_TENSOR_TYPE_ASSOC error in test_util.h file -- 2ff70a8 by RahulSudarMCW <rahul.sundar@multicorewareinc.com>: Include non-quantized int8 & int16 type support for tfl.pad Merging this change closes #74506 FUTURE_COPYBARA_INTEGRATE_REVIEW=#74506 from MCW-Dev:master 2ff70a8 PiperOrigin-RevId: 750807771

* TfLite Reverse missing datatype support -Adds bf16, f16 support for reverse -Adds bf16, f16 reverse unit tests

spk921 closed this as completed Nov 13, 2015

bluezone2015 mentioned this issue Mar 8, 2017

import tensorflow Segmentation fault (core dumped) #8197

Closed

zhangdingfei mentioned this issue Jun 21, 2017

tensorflow-1.2.0 import tensorflow Segmentation fault #10870

Closed

ndvbd mentioned this issue Jul 19, 2018

Build tensorflow-model-server for gpu - Cannot find cuda library libcudnn.so.6 #20959

Closed

EvansDaniel mentioned this issue Sep 8, 2018

Unexpected behavior of per_process_gpu_memory_fraction #22158

Closed

bhagatindia mentioned this issue Sep 11, 2018

Session run options for configuring the timeout of a run is not working as expected #22217

Closed

chenjiasheng mentioned this issue Dec 12, 2018

Distributed Training Randomly Stops During the Training Process #12667

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

cjolivier01 pushed a commit to Cerebras/tensorflow that referenced this issue Dec 6, 2019

Merge pull request tensorflow#71 from ROCmSoftwarePlatform/refactor-a…

008b9db

…lgorithmconfig-profileresult Refactor AlgorithmConfig / ProfileResult and put scratch size in AlgorithmDesc

craterkamath mentioned this issue Jan 16, 2020

Install Tensorflow 1.15 on CUDA 10.2 and cudnn 7.6 #35926

Closed

copybara-service bot pushed a commit that referenced this issue May 17, 2022

Add security advisory for vuln #71 in 2022

65bff07

PiperOrigin-RevId: 449112147

ramonhollands mentioned this issue Jun 30, 2024

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

Closed

copybara-service bot mentioned this issue Mar 4, 2025

PR #74506: Adds missing datatype support for various tflite operations #88580

Draft

copybara-service bot mentioned this issue Apr 24, 2025

PR #74506: Adds missing datatype support for various tflite operations #92093

Merged

copybara-service bot pushed a commit that referenced this issue Apr 28, 2025

TfLite Reverse missing datatype support (#31) (#71)

baea4f1

* TfLite Reverse missing datatype support -Adds bf16, f16 support for reverse -Adds bf16, f16 reverse unit tests

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Uh oh!

Can't install from source if I don't have a GPU? #71

Can't install from source if I don't have a GPU? #71

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Can't install from source if I don't have a GPU? #71

Can't install from source if I don't have a GPU? #71

Comments

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!