I have a symmetrc matrix with some parameters to be changed. For some parameters, the LAPACKE_dsyevr does not return for a very long time, so I have to finish the program.
If I change to LAPACKE_dsyev, for most cases, the speed is slower than dsyevr, but the dsyev can return correctly.
What is the reason?
Thanks for your help.
The dimension of matrix is 10000.
In addition , I still have a question. When we compute the symmetric or Hermitian problem, is it necessary to set the lower part to be all zeros?
If we use the whole matrix as input with both upper and lower part, the result will be wrong?
Gennady F. (Intel) wrote:
What is the problem size when you are calling the dsyevr and finish this execution?