search

bpartridge / PyPhi

Python for Scientific Computing on Intel's Xeon Phi MIC Platform
22 stars 7 forks source link

patch: **** Only garbage was found in the patch input. #5

Open jinhoonbang opened 8 years ago

jinhoonbang commented 8 years ago

I'm trying to compile Numpy and Scipy on Xeon Phi. While running the Makefile, I got an error saying patch: **\ Only garbage was found in the patch input.

From reading the log, the error occurs when patch is applied by the command : "cd /home/dkl524/jbang/mic/python && patch -p1 < ../xcompile.patch"

Does anyone know solution to this?

jinhoonbang commented 8 years ago

Realized that the website (randomsplat), from which the patch is downloaded, wasn't working correctly. Trying to run make file using a different patch file now.

jinhoonbang commented 8 years ago

I get following error when running the command: cd numpy && PYTHONXCPREFIX=$(MIC_PY_HOME) python numpyxc.py build"

icc -mmic -mkl -fPIC -fp-model strict -g -I/home/dkl524/jbang/mic/python/_install/include -L/home/dkl524/jbang/mic/python/_install/lib -L/home/dkl524/jbang/mic/python/_install/lib/python2.7 -L/home/dkl524/jbang/mic/python/_install/lib/python2.7/site-packages/numpy/core/lib -shared build/temp.linux-x86_64-2.6/numpy/core/src/dummymodule.o -L/usr/lib64 -Lbuild/temp.linux-x86_64-2.6 -lpython2.6 -o build/lib.linux-x86_64-2.6/numpy/core/_dummy.so icc: command line remark #10364: Using '-fp-model strict' forces the compiler to generate legacy floating-point instructions instead of KNC for most floating-point operations, which may impact performance. Consider using a floating-point model that allows the compiler to assume the default floating-point environment. x86_64-k1om-linux-ld: skipping incompatible /usr/lib64/libpython2.6.so when searching for -lpython2.6 x86_64-k1om-linux-ld: cannot find -lpython2.6 icc: command line remark #10364: Using '-fp-model strict' forces the compiler to generate legacy floating-point instructions instead of KNC for most floating-point operations, which may impact performance. Consider using a floating-point model that allows the compiler to assume the default floating-point environment. x86_64-k1om-linux-ld: skipping incompatible /usr/lib64/libpython2.6.so when searching for -lpython2.6 x86_64-k1om-linux-ld: cannot find -lpython2.6 error: Command "icc -mmic -mkl -fPIC -fp-model strict -g -I/home/dkl524/jbang/mic/python/_install/include -L/home/dkl524/jbang/mic/python/_install/lib -L/home/dkl524/jbang/mic/python/_install/lib/python2.7 -L/home/dkl524/jbang/mic/python/_install/lib/python2.7/site-packages/numpy/core/lib -shared build/temp.linux-x86_64-2.6/numpy/core/src/dummymodule.o -L/usr/lib64 -Lbuild/temp.linux-x86_64-2.6 -lpython2.6 -o build/lib.linux-x86_64-2.6/numpy/core/_dummy.so" failed with exit status 1 make: *\ [numpyxc] Error 1

Why is it looking for python 2.6 all of a sudden. I thought the makefile is supposed to install 2.7.3