DenseMatrixInversionResultsinSegmentationFault:CausesandSolutions

Expected behaviour

smile> val a = randn(100,100)
a: DenseMatrix = -0.2015 0.0481 0.0432 -0.6100 -1.0286 -0.7054 0.1032 ...
0.9704 0.3677 -0.0011 1.5735 -0.7982 1.1519 -0.3758 ...
2.7269 -0.7253 -0.2067 -0.8212 1.8176 -0.7403 -0.1626 ...
-0.1460 -1.6094 1.0685 -0.7500 -2.9544 0.9874 -0.7922 ...
1.8722 0.6541 -1.5475 1.7311 1.5635 1.2492 -0.6935 ...
0.4959 0.6218 1.8119 2.1485 0.0260 -0.2688 0.0886 ...
0.5539 -1.8473 0.8755 -0.2812 1.3386 -0.3521 0.3232 ...
...

The expected output is the matrix inverse of DenseMatrix a.

Actual behaviour

smile> val b = inv(a)
Nov 03, 2017 1:13:28 AM com.github.fommil.jni.JniLoader liberalLoad
INFO: successfully loaded /tmp/jniloader7844851871842582915netlib-native_system-linux-x86_64.so
Segmentation fault

Code snippet

val a = randn(100,100)
val b = inv(a)

Input data

randn(100,100)

Information

• What Java (OpenJDK, Orack JDK, etc.) are you using and which Java version
  openjdk 1.8.0_131




• Which Smile version
  1.5.0 RC2




• What is your build system (e.g. Ubuntu, MacOS, Windows, Debian )
  Ubuntu 16.04.3 LTS x64 32Gb RAM i7 6700

该提问来源于开源项目：haifengl/smile