topic Yes sorry, in fact i had in Intel® oneAPI Math Kernel Library

Bug dgeqrf mkl_sequential

pat97269 — Mon, 01 Oct 2012 23:45:10 GMT

Hi,

There is a problem with the dgeqrf function in sequential with the last mkl. The intel_thread give me the same results as lapack 3.4.1,, but in the sequential version the matrix seems

unchanged.

I provide a small program and data file for you to confirm that.

I use : ifort testdgeqrf.f90 -lmkl_sequential -lmkl_core -lmkl_intel_lp64

the test requires DDG.txt

Gennady_F_Intel — Tue, 02 Oct 2012 04:02:47 GMT

the test requires DDG.txt file. e.g: open(10,file='DDG.txt') I see - ddg.log is attached. Is this those file?

Yes sorry, in fact i had

pat97269 — Tue, 02 Oct 2012 12:51:55 GMT

Yes sorry, in fact i had trouble uploading a *.txt file, so i renamed it .log . It contains the matrix that fails for me ! Thanks