3

I tried using IMKL through EIGEN by enabling EIGEN_USE_MKL_ALL flag. But when i compared the assembly code of 1. MKL alone 2. enabling EIGEN_USE_MKL_ALL using EIGEN

both are same...

shakkeeb M
  • 31
  • 2
  • 1
    Please show the exact expression you tried, not all operations are mapped to MKL. – ggael Oct 05 '15 at 09:00
  • #include #define EIGEN_USE_MKL_ALL #include "Eigen/Dense" using Eigen::MatrixXd; int main() { MatrixXd m1,m2,m3; m1<<1,2,3,4; m2<<5,6,7,8; m3=m1*m2.Transpose(); std::cout< – shakkeeb M Oct 12 '15 at 11:37
  • Possible duplicate of [how to invoke IMKL functions from Eigen](http://stackoverflow.com/questions/33080650/how-to-invoke-imkl-functions-from-eigen) – ggael Oct 12 '15 at 13:43
  • Hi, I have a similar problem http://stackoverflow.com/questions/40868462/no-speed-up-after-using-mkl-for-eigen?noredirect=1#comment68968960_40868462 . How did you solve your problem? – olivia Nov 30 '16 at 03:37

0 Answers0