[Numpy-discussion] dot and MKL memory

Daπid davidmenhur at gmail.com
Fri Feb 27 12:43:12 EST 2015


On 27 February 2015 at 17:50, Charles R Harris
<charlesr.harris at gmail.com> wrote:
>>
>
> No, but I don't know why that is happening with MKL. Can anyone else
> reproduce this?
>
> Chuck

I can't reproduce. I have checked on my system python (ATLAS) and
Conda with MKL running in parallel with vmstat. The difference between
them is under 1 MB.

I am Fedora 20, system python using GCC 4.8 and ATLAS 3.8.4. Coda is
compiled with GCC 4.4 and linked to MKL 11.1.


/David.



More information about the NumPy-Discussion mailing list