benoitsteiner / tensorflow-xsmm Goto Github PK
View Code? Open in Web Editor NEWImproved performance for TensorFlow on Intel hardware.
License: Apache License 2.0
Improved performance for TensorFlow on Intel hardware.
License: Apache License 2.0
When mkl_include symlink already exists, the configure script tries to write the new include symlink in the directory pointed to by mkl_include. This is easily fixed by adding the -n option to ln.
This is also present in the current tensorflow master branch, but I think this is the upstream.
diff --git a/configure b/configure
index 23603a9..7605a2c 100755
--- a/configure
+++ b/configure
@@ -252,14 +252,14 @@ if [ "$TF_NEED_MKL" == "1" ]; then # TF_NEED_MKL
ln -sf $MKL_INSTALL_PATH/${MKL_ML_LIB_PATH} third_party/mkl/
ln -sf $MKL_INSTALL_PATH/${MKL_ML_OMP_LIB_PATH} third_party/mkl/
ln -sf $MKL_INSTALL_PATH/include third_party/mkl/
- ln -sf $MKL_INSTALL_PATH/include third_party/eigen3/mkl_include
+ ln -sfn $MKL_INSTALL_PATH/include third_party/eigen3/mkl_include
loc=$(locate -e libdl.so.2 | sed -n 1p)
ln -sf $loc third_party/mkl/libdl.so.2
elif [ -e "$MKL_INSTALL_PATH/${MKL_RT_LIB_PATH}" ]; then
ln -sf $MKL_INSTALL_PATH/${MKL_RT_LIB_PATH} third_party/mkl/
ln -sf $MKL_INSTALL_PATH/${MKL_RT_OMP_LIB_PATH} third_party/mkl/
ln -sf $MKL_INSTALL_PATH/include third_party/mkl/
- ln -sf $MKL_INSTALL_PATH/include third_party/eigen3/mkl_include
+ ln -sfn $MKL_INSTALL_PATH/include third_party/eigen3/mkl_include
loc=$(locate -e libdl.so.2 | sed -n 1p)
ln -sf $loc third_party/mkl/libdl.so.2
else
There are any differences or custom flags from building cpu-only common tensorflow?
Thanks in advance!
A declarative, efficient, and flexible JavaScript library for building user interfaces.
๐ Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
The Web framework for perfectionists with deadlines.
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. ๐๐๐
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
We are working to build community through open source technology. NB: members must have two-factor auth.
Open source projects and samples from Microsoft.
Google โค๏ธ Open Source for everyone.
Alibaba Open Source for everyone
Data-Driven Documents codes.
China tencent open source team.