transfer /opt/local to another machine
jonesc at hep.phy.cam.ac.uk
Mon Apr 13 02:34:29 PDT 2015
On 13/04/15 10:27, j. van den hoff wrote:
> thanks to both of you for your answers and clarifications.
> On Mon, 13 Apr 2015 04:04:14 +0200, Ryan Schmidt
> <ryandesign at macports.org> wrote:
>> On Apr 12, 2015, at 5:54 AM, j. van den hoff wrote:
>>> I just have upgraded two x86_64 machines from 10.9. to 10.10.3 and
>>> upgraded one of them according to the migration guide (i.e.
>>> uninstalled/reinstalled all ~ 800 macports packages). this took an
>>> astonishing long time (over night, basically) partly due to a
>>> multi-hour recompilation of the `atlas' library. and that was on the
>>> faster machine with an SSD...
>> This is intentional, because atlas is performance-sensitive, so the
>> port is deliberately programmed to not use precompiled binaries from
>> our build server, and to instead build on your computer. My
>> understanding is that the build process builds atlas many times, with
>> different compiler settings, then tries out each one to see which one
>> is actually fastest on your particular computer, then installs that
>> one. That's why it takes so long.
> yes, I understand that. I'm not sure, however, how much of a performance
> hit I will see when using the intel core-i7 optimized binary on an intel
> core 2 duo, so I just would try it out, whether it's OK for my use (that
> would mean via `octave', which I am using only sporadically, so ...).
If the core-i7 build is using CPU instructions your core 2 due does not
have, then the performance hit will be 'significant' (i.e. it simply
will not run...).
More information about the macports-users